A dietary intervention management method for children with renal disease undergoing hemodialysis

By obtaining and pre-processing the personal information data of children with hemodialysis of kidney disease, performing cluster analysis and particle swarm optimization, the lack of personalization and flexibility in traditional diet management methods is solved, and a personalized and real-time adjustment of diet plan is achieved. Improves treatment effect and quality of life.

CN118692629BActive Publication Date: 2024-11-19BEIJING SIHAI HUIZHI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411195598.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-19
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

The traditional dietary management method of children with hemodialysis patients with kidney disease fails to fully consider and deeply analyze the patient's personal information data, resulting in a lack of personalization and flexibility in diet plans, affecting treatment results and quality of life.

Method used

By obtaining the patient's personal information data, performing preprocessing and clustering analysis, the patients are divided into different groups, initializing the particle swarm and iteratively optimized using the fitness function, setting a dynamic threshold set, and formulating a personalized diet plan.

Benefits of technology

It has achieved more comprehensive and personalized diet management, improved data processing efficiency and treatment effect, and can adjust diet plans in real time according to the actual situation of the patients, improving the quality of life.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118692629B_ABST
    Figure CN118692629B_ABST
Patent Text Reader

Abstract

The present invention relates to a dietary intervention management method for children with renal disease hemodialysis, and relates to the field of data processing technology. The method comprises: pre-processing the personal information data to obtain pre-processed data; performing cluster analysis according to the pre-processed data, and dividing the patients into different groups to obtain clustering results; setting an initial dietary intake index threshold for each group according to the clustering results to form an initial dynamic threshold set; initializing a particle group, each particle representing a set of thresholds, and determining a fitness function for evaluating the gap between the actual blood parameters of the patient and the target value; according to the fitness function, iteratively updating the speed and position of the particles to find the corresponding threshold, and when the preset number of iterations is reached, the final dynamic threshold set is obtained. The present invention can divide patients into different groups according to the similarity of their personal information, thereby improving the efficiency of data processing.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a dietary intervention management method for children with renal disease and hemodialysis. Background Art

[0002] Diet management is a crucial link in the treatment of pediatric renal hemodialysis patients. Although traditional dietary intervention methods have accumulated certain experience in practice, they still need to be improved in data processing.

[0003] Specifically, traditional methods sometimes fail to fully consider and deeply analyze patients' personal information data, such as age, gender, height, weight, hemodialysis frequency, etc. These data are critical for accurately formulating a diet plan that meets the individual needs of patients. In addition, the integration of information such as the patient's medical history, eating habits and preferences can also help improve the acceptance and implementation of the diet plan.

[0004] Furthermore, traditional diet management methods have certain limitations in terms of dynamic adjustment. It may lack scientific data analysis methods, making it difficult to adjust diet plans in real time according to changes in patients' biochemical indicators and physical conditions. This lack of flexibility may affect the treatment effect and the quality of life of patients. Therefore, when faced with a large amount of patient data, the processing efficiency of traditional methods may not be satisfactory, so improvement is urgently needed. 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 dietary intervention management method for pediatric renal hemodialysis patients, which can divide patients into different groups according to the similarity of their personal information, thereby improving the efficiency of data processing.

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

[0007] In the first aspect, a dietary intervention management method for pediatric renal hemodialysis patients comprises:

[0008] Obtain personal information data of pediatric renal hemodialysis patients, including age, gender, height, weight, hemodialysis frequency, medical history, dietary habits and preferences;

[0009] Preprocessing the personal information data to obtain preprocessed data;

[0010] Performing cluster analysis based on the preprocessed data and dividing the patients into different groups to obtain clustering results;

[0011] According to the clustering results, the initial dietary intake index threshold is set for each group to form an initial dynamic threshold set;

[0012] Initialize a particle swarm, each particle represents a set of thresholds, and determine a fitness function for evaluating the gap between the actual blood parameter of the patient and the target value;

[0013] According to the fitness function, the particle speed and position are updated iteratively to find the corresponding threshold. When the preset number of iterations is reached, the final dynamic threshold set is obtained.

[0014] According to the final dynamic threshold set, the actual data of each patient is compared with the dynamic threshold to obtain a comparison result;

[0015] A diet plan is developed for the patient based on the comparison results and the patient's specific situation.

[0016] Further, the personal information data is preprocessed to obtain preprocessed data, including:

[0017] Personal information data Performing filtering processing to obtain filtered data;

[0018] in, represents the filtered data, N represents the window size of the moving average, and represents the number of past data points when calculating the average; represents the weight coefficient; Indicates the input signal at time point The value of represents the modulation term of the sinusoidal waveform, is the amplitude, is the angular frequency, is the index of the data point; represents the modulation term of the cosine waveform, is the amplitude, is the angular frequency, is the index of the current time point; represents the attenuation factor;

[0019] The filtered data is processed to obtain preprocessed data.

[0020] Further, the filtered data is processed to obtain preprocessed data, including:

[0021] The filtered data is passed Processing is performed to obtain preprocessed data, where represents the preprocessed data, which represents the normalized output value at time point n; and are the maximum and minimum values ​​in the filtered data set, respectively; represents the scaling factor; Represents the bias term.

[0022] Furthermore, cluster analysis is performed based on the preprocessed data, and the patients are divided into different groups to obtain clustering results, including:

[0023] Determine the number of cluster groups k, and determine k data points from the preprocessed data as the initial cluster centers;

[0024] For each data point in the preprocessed data, calculate the distance between it and each cluster center, assign each data point to the corresponding cluster center, and form k cluster center groups;

[0025] For each cluster center group, the average value of all data points in the group is calculated, the average value is used as the new cluster center, the data points are redistributed to the nearest cluster center, and the position of the cluster center is updated until the preset number of iterations is reached to obtain the final clustering result.

[0026] Furthermore, the fitness function The calculation formula is:

[0027] ;

[0028] in, Indicates Actual measured values ​​of blood parameters; Indicates target values ​​for blood parameters; Indicates The weight of each blood parameter; , and is a hyperparameter; represents the base of natural logarithms; represents the weight of the cumulative effect term; Indicates and The correlation coefficient between the blood parameters; m represents the number of parameter combinations; Indicates the total number of blood parameters; and They represent different blood parameters respectively; The value range is 1 to ; The value range is 1 to .

[0029] Furthermore, when updating the velocity and position of a particle, the velocity update formula is:

[0030] ;

[0031] in, It is a particle In dimension d The speed of the moment; It is a particle The velocity at time t in dimension d; B is the inertia weight; represents the individual learning factor, represents the social learning factor; and is a random number in the range [0, 1]; It is a particle The optimal position of an individual on dimension d; is the global optimal position of the entire particle swarm in dimension d; It is a particle The position at time t in dimension d.

[0032] Furthermore, when updating the velocity and position of the particle, the update formula of the position is:

[0033] ;

[0034] in, It is a particle In dimension d The location at the moment; It is a particle In dimension d The speed of time.

[0035] In the second aspect, a dietary intervention management system for children with renal hemodialysis includes:

[0036] An acquisition module is used to acquire personal information data of pediatric renal hemodialysis patients, including age, gender, height, weight, hemodialysis frequency, medical history, dietary habits and preferences; preprocess the personal information data to obtain preprocessed data; perform cluster analysis based on the preprocessed data and divide the patients into different groups to obtain clustering results; and set an initial dietary intake index threshold for each group based on the clustering results to form an initial dynamic threshold set;

[0037] The processing module is used to initialize the particle group, each particle represents a set of thresholds, and determine the fitness function used to evaluate the gap between the patient's actual blood parameters and the target values; according to the fitness function, the speed and position of the particles are iteratively updated to find the corresponding thresholds, and when the preset number of iterations is reached, the final dynamic threshold set is obtained; according to the final dynamic threshold set, the actual data of each patient is compared with the dynamic threshold to obtain the comparison result; according to the comparison result and the specific situation of the patient, a diet plan is formulated for the patient.

[0038] According to a third aspect, a computing device includes:

[0039] one or more processors;

[0040] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0041] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0042] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0043] The system comprehensively collects and integrates the patient's multi-dimensional personal information data, including age, gender, height, weight, hemodialysis frequency, medical history, eating habits and preferences, etc. By integrating these data, we can have a more comprehensive understanding of the patient's physical condition and living habits.

[0044] Preprocessing the collected raw data improves the quality and availability of the data. Clustering analysis of patient data using clustering algorithms can divide patients into different groups based on their similarities in personal information. Based on the clustering results, an initial dietary intake index threshold is set for each group, and a dynamic threshold set is formed. This dynamic setting method can be flexibly adjusted according to the actual situation of the patient.

[0045] By initializing the particle swarm and using the fitness function for iterative optimization, this method can quickly find the optimal threshold combination and improve the efficiency of data processing.

[0046] According to the final dynamic threshold set, the patient's actual data is compared with the dynamic threshold, which can monitor the patient's dietary intake in real time and adjust the dietary plan in time according to the comparison results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. Some specific embodiments of the present application will be described in detail in an illustrative and non-restrictive manner with reference to the drawings. The same reference numerals in the drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings:

[0048] Figure 1 The present invention is a flowchart of a method for dietary intervention management for children with renal disease undergoing hemodialysis.

[0049] Figure 2It is a schematic diagram of a dietary intervention management system for children with renal disease undergoing hemodialysis according to the present invention. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0051] The following embodiment of the present application takes the dietary intervention management method for children with renal disease hemodialysis as an example to explain the scheme of the present application in detail, but this embodiment cannot limit the scope of protection of the present application.

[0052] like Figure 1 As shown, the present invention provides a method for dietary intervention management of children with renal disease and hemodialysis, the method comprising the following steps:

[0053] Step 11, obtaining personal information data of pediatric renal hemodialysis patients, including age, gender, height, weight, hemodialysis frequency, medical history, dietary habits and preferences;

[0054] Step 12, preprocessing the personal information data to obtain preprocessed data;

[0055] Step 13, performing cluster analysis based on the preprocessed data, and dividing the patients into different groups to obtain clustering results;

[0056] Step 14, according to the clustering results, an initial dietary intake index threshold is set for each group to form an initial dynamic threshold set;

[0057] Step 15, initializing a particle swarm, each particle representing a set of thresholds, and determining a fitness function for evaluating the gap between the actual blood parameter of the patient and the target value;

[0058] Step 16, according to the fitness function, the speed and position of the particle are updated through iteration to find the corresponding threshold value, and when the preset number of iterations is reached, the final dynamic threshold value set is obtained;

[0059] Step 17, according to the final dynamic threshold set, the actual data of each patient is compared with the dynamic threshold to obtain a comparison result;

[0060] Step 18, formulating a diet plan for the patient based on the comparison results and the patient's specific conditions.

[0061] In an embodiment of the present invention, the method first collects comprehensive personal information data of the patient, including age, gender, height, weight, etc. This data-driven method can ensure that each patient can get a customized diet plan based on their personal characteristics, greatly improving the pertinence and effectiveness of dietary intervention. By preprocessing the original personal information data, including filtering, normalization and other operations, the method can eliminate noise and outliers in the data. The optimized data quality helps to improve the accuracy of cluster analysis, thereby more accurately dividing patient groups. Cluster analysis using preprocessed data can divide patients with similar characteristics into the same group, which helps to identify the common characteristics of different patient groups; by setting an initial dietary intake index threshold for each patient group and forming an initial dynamic threshold set, the method can adjust the dietary management strategy according to the characteristics of different groups, and can better adapt to individual differences and changes in the condition of patients. Through the particle swarm optimization algorithm, the method can efficiently search for the optimal dietary intake index threshold, improving the search efficiency and accuracy. By comparing the actual data of the patient with the dynamic threshold in real time, the method can promptly discover problems and make adjustments. This real-time feedback mechanism helps to ensure the effectiveness of dietary intervention and promptly correct dietary behaviors that deviate from the target.

[0062] In a further embodiment of the present invention, the above step 12, preprocessing the personal information data to obtain preprocessed data, may include:

[0063] Step 121, personal information data is processed through Performing filtering processing to obtain filtered data;

[0064] in, represents the filtered data, N represents the window size of the moving average, and represents the number of past data points when calculating the average; represents the weight coefficient; Indicates the input signal at time point The value of represents the modulation term of the sinusoidal waveform, is the amplitude, is the angular frequency, is the index of the data point; represents the modulation term of the cosine waveform, is the amplitude, is the angular frequency, is the index of the current time point; represents the attenuation factor;

[0065] Step 122, processing the filtered data to obtain preprocessed data.

[0066] In the embodiment of the present invention, the noise and abnormal fluctuations in the original data can be effectively suppressed by filtering, making the data smoother. The sine and cosine waveform modulation terms introduced in the filtering formula can enhance the periodic characteristics of the signal to a certain extent. , data at different time points can be weighted, and its influence can be adjusted according to the timeliness and importance of the data, so that more recent data has a greater weight in the preprocessing process, thereby improving the timeliness and relevance of the data. The filtering formula contains multiple adjustable parameters, such as window size N, attenuation factor ,amplitude and , angular frequency and Etc. These parameters can be adjusted according to specific data characteristics and analysis requirements, so as to ensure that the preprocessing process can better adapt to different data sets and analysis scenarios. When clustering analysis is performed on preprocessed data, the accuracy of clustering can be significantly improved due to the reduction of noise and outliers and the enhancement of signal characteristics, which means that patients can be more accurately divided into different groups.

[0067] In a further embodiment of the present invention, the above step 121, processing the filtered data to obtain pre-processed data, may include:

[0068] Step 1211, filter the data by Processing is performed to obtain preprocessed data, where represents the preprocessed data, which represents the normalized output value at time point n; and are the maximum and minimum values ​​in the filtered data set, respectively; represents the scaling factor; Represents the bias term.

[0069] In an embodiment of the present invention, the normalization process can limit the range of the filtered data to between 0 and 1, realize the standardization of the data, help eliminate the influence of the original scale or unit of the data, and make the data of different dimensions comparable in numerical value. The normalized data can be used as input to improve the performance of many machine learning algorithms. Because many algorithms are sensitive to the scale of the input data, normalization can ensure that the algorithm works more stably and reduce the performance degradation caused by the difference in data scale. By limiting the data to a reasonable range, it can be ensured that the network can learn the characteristics of the data more effectively during the training process. Normalization can make the optimization algorithm (such as gradient descent) converge faster when looking for the optimal solution. Because the data is limited to a relatively small range, the optimization algorithm can approach the optimal solution faster in each iteration. The normalized data is easier to visualize because they are all on a unified scale, and normalization can reduce the impact of outliers on the data set because all data are mapped to the same range. This enhances the robustness of the data set and makes the analysis results more stable and reliable.

[0070] In a further embodiment of the present invention, the above step 13, performing cluster analysis based on the pre-processed data and dividing the patients into different groups to obtain clustering results, may include:

[0071] Step 131, determining the number of cluster groups k, and determining k data points from the preprocessed data as initial cluster centers;

[0072] Step 132, for each data point in the preprocessed data, calculate the distance between it and each cluster center, assign each data point to the corresponding cluster center, and form k cluster center groups;

[0073] Step 133, for each cluster center group, calculate the average value of all data points in the group, use the average value as the new cluster center, redistribute the data points to the nearest cluster center, and update the position of the cluster center until the preset number of iterations is reached to obtain the final clustering result.

[0074] In an embodiment of the present invention, k data points are randomly or purposefully selected from the preprocessed data as initial cluster centers. These points will serve as the starting point of the clustering algorithm; for each data point in the preprocessed data, the Euclidean distance between it and the current k cluster centers is calculated; each data point is assigned to the nearest cluster center, thereby forming k preliminary cluster groups; for each cluster group, the average value of all data points in it is calculated, and this average value is set as the new cluster center; according to the new cluster center, all data points are reallocated to the nearest cluster center; the steps are repeated, that is, the cluster center of each group is recalculated, and the position of the cluster center is updated. This process is repeated until the preset number of iterations is reached; after the algorithm converges, the final cluster center and the group affiliation of each data point are output as the result of cluster analysis. In an embodiment of the present invention, similar patient data can be grouped together through cluster analysis, which helps to understand and discover patterns and structures in the data.

[0075] In a further embodiment of the present invention, the calculation formula of the fitness function is:

[0076] ;

[0077] in, Indicates Actual measured values ​​of blood parameters; Indicates target values ​​for blood parameters; Indicates The weight of each blood parameter; , and is a hyperparameter; represents the base of natural logarithms; represents the weight of the cumulative effect term; Indicates and The correlation coefficient between the blood parameters; m represents the number of parameter combinations; Indicates the total number of blood parameters; and They represent different blood parameters respectively; The value range is 1 to ; The value range is 1 to .

[0078] In the embodiment of the present invention, by weight , different importance can be assigned to different blood parameters; the function not only considers the absolute difference between the actual measured value and the target value, but also introduces square error terms and exponential error terms, which makes the error metric more comprehensive and can better punish large deviations. By considering the correlation between parameters (represented by the combination of m parameters), the function can better understand and utilize the mutual influence between parameters, thereby improving the accuracy and efficiency of optimization.

[0079] In a further embodiment of the present invention, when updating the velocity and position of a particle, the velocity update formula is:

[0080] ;

[0081] in, It is a particle In dimension d The speed of the moment; It is a particle The velocity at time t in dimension d; B is the inertia weight; represents the individual learning factor, represents the social learning factor; and is a random number in the range [0, 1]; It is a particle The optimal position of an individual on dimension d; is the global optimal position of the entire particle swarm in dimension d; It is a particle The position at time t in dimension d; when updating the velocity and position of the particle, the update formula of the position is:

[0082] ;

[0083] in, It is a particle In dimension d The location at the moment; It is a particle In dimension d The speed of time.

[0084] In the embodiment of the present invention, the global search capability and the local search capability of the algorithm can be balanced by adjusting the inertia weight B. A larger B value is conducive to global search, while a smaller B value strengthens local search. and Represents the individual learning factor and social learning factor, respectively, which control the degree to which particles learn from the individual optimal solution and the global optimal solution. and The introduction of increases the randomness of the search process, helps the algorithm to jump out of the local optimal solution and increases the chance of finding the global optimal solution.

[0085] In step 16, after each iteration, the fitness value of each particle is recalculated, and the new fitness value of each particle is compared with the fitness values ​​of its individual optimal solution and the global optimal solution, and updated accordingly; the steps are repeated until the preset number of iterations is reached. After the iteration is completed, the global optimal solution is output as the final dynamic threshold set.

[0086] In step 17, actual physiological data of each patient is collected, such as blood sugar, blood pressure, heart rate, etc.; necessary preprocessing is performed on the patient data, such as cleaning, standardization or normalization, to ensure that the data format matches the dynamic threshold set; the preprocessed patient data is compared with the dynamic threshold set obtained in step 16 to check whether the data points fall within a specific threshold range; based on the comparison, the comparison results of each patient data point with the dynamic threshold are recorded, for example, which indicators are within the normal range and which exceed the threshold.

[0087] In step 18, based on the comparison results in step 17, analyze the patient's physiological indicators and identify which indicators are abnormal or require special attention. In addition to physiological data, the patient's personal conditions, such as age, gender, weight, activity level, eating habits, preferences, and any known health problems or allergies, should also be considered. Based on the patient's physiological data and personal conditions, a preliminary diet plan is developed. This plan includes recommended food types, portions, eating frequency, and foods to avoid. Communicate with patients and adjust and optimize the diet plan based on their feedback and actual conditions to ensure the feasibility of the plan and patient acceptance. After implementing the diet plan, continue to monitor changes in the patient's physiological indicators and make corresponding adjustments as needed to achieve reasonable health results.

[0088] Figure 2 As shown, the embodiment of the present invention also provides a dietary intervention management system 20 for children with renal disease and hemodialysis, comprising:

[0089] The acquisition module 21 is used to obtain personal information data of pediatric renal hemodialysis patients, including age, gender, height, weight, hemodialysis frequency, medical history, dietary habits and preferences; pre-process the personal information data to obtain pre-processed data; perform cluster analysis based on the pre-processed data and divide the patients into different groups to obtain clustering results; and set an initial dietary intake index threshold for each group based on the clustering results to form an initial dynamic threshold set;

[0090] The processing module 22 is used to initialize the particle group, each particle represents a set of thresholds, and determine the fitness function used to evaluate the gap between the patient's actual blood parameters and the target values; according to the fitness function, the speed and position of the particles are iteratively updated to find the corresponding thresholds, and when the preset number of iterations is reached, the final dynamic threshold set is obtained; according to the final dynamic threshold set, the actual data of each patient is compared with the dynamic threshold to obtain a comparison result; according to the comparison result and the specific situation of the patient, a diet plan is formulated for the patient.

[0091] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0092] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0093] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0094] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A dietary intervention management method for children with renal disease undergoing hemodialysis, characterized in that: include: Obtain personal information data of pediatric renal hemodialysis patients, including age, gender, height, weight, hemodialysis frequency, medical history, dietary habits and preferences; Preprocessing the personal information data to obtain preprocessed data; Performing cluster analysis based on the preprocessed data and dividing the patients into different groups to obtain clustering results; According to the clustering results, the initial dietary intake index threshold is set for each group to form an initial dynamic threshold set; Initialize a particle swarm, each particle represents a set of thresholds, and determine a fitness function for evaluating the gap between the actual blood parameter of the patient and the target value; According to the fitness function, the particle speed and position are updated iteratively to find the corresponding threshold. When the preset number of iterations is reached, the final dynamic threshold set is obtained. According to the final dynamic threshold set, the actual data of each patient is compared with the dynamic threshold to obtain a comparison result; Develop a diet plan for the patient based on the comparison results and the patient's specific situation; Preprocessing the personal information data to obtain preprocessed data includes: Personal information data Performing filtering processing to obtain filtered data; in, represents the filtered data, N represents the window size of the moving average, and represents the number of past data points when calculating the average; represents the weight coefficient; Indicates the input signal at time point The value of represents the modulation term of the sinusoidal waveform, is the amplitude, is the angular frequency, is the index of the data point; represents the modulation term of the cosine waveform, is the amplitude, is the angular frequency, is the index of the current time point; represents the attenuation factor; Processing the filtered data to obtain preprocessed data; The filtered data is processed to obtain preprocessed data, including: The filtered data is passed Processing is performed to obtain preprocessed data, where represents the preprocessed data, which represents the normalized output value at time point n; and are the maximum and minimum values ​​in the filtered data set, respectively; represents the scaling factor; represents the bias term; Cluster analysis is performed based on the pre-processed data, and the patients are divided into different groups to obtain clustering results, including: Determine the number of cluster groups k, and determine k data points from the preprocessed data as the initial cluster centers; For each data point in the preprocessed data, calculate the distance between it and each cluster center, assign each data point to the corresponding cluster center, and form k cluster center groups; For each cluster center group, calculate the average value of all data points in the group, use the average value as the new cluster center, redistribute the data points to the nearest cluster center, and update the position of the cluster center until the preset number of iterations is reached to obtain the final clustering result; Fitness function The calculation formula is: ; in, Indicates Actual measured values ​​of blood parameters; Indicates target values ​​for blood parameters; Indicates The weight of each blood parameter; , and is a hyperparameter; represents the base of natural logarithms; represents the weight of the cumulative effect term; Indicates and The correlation coefficient between the blood parameters; m represents the number of parameter combinations; Indicates the total number of blood parameters; and They represent different blood parameters respectively; The value range is 1 to ; The value range is 1 to ; When updating the velocity and position of a particle, the velocity update formula is: ; in, It is a particle In dimension d The speed of the moment; It is a particle The velocity at time t in dimension d; B is the inertia weight; represents the individual learning factor, represents the social learning factor; and is a random number in the range [0, 1]; It is a particle The optimal position of an individual on dimension d; is the global optimal position of the entire particle swarm in dimension d; It is a particle The position at time t in dimension d; When updating the velocity and position of a particle, the update formula for the position is: ; in, It is a particle In dimension d The location at the moment; It is a particle In dimension d The speed of time.

2. A computing device, characterized in that: include: one or more processors; A storage device, used for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in claim 1.

3. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which implements the method according to claim 1 when executed by a processor.