Physical examination data processing and analyzing method and system

Abstract

The invention relates to the technical field of data analysis, in particular to a physical examination data processing analysis method and system, and the method comprises the following steps: extracting individual health indexes from a physical examination data set, classifying the health indexes according to types and time, comparing real-time physical examination data with historical physical examination data, and generating differentiated health index data. According to the method, the physical examination data are comprehensively analyzed, detailed classification and time sequence comparison of individual health indexes are achieved, the difference between real-time data and historical data is effectively revealed, risk factors are further graded through detailed analysis of periodic modes and aperiodic mutation of abnormal events, and the risk factors are classified to be more accurate. According to the method, more personalized health risk assessment is provided, future health events are predicted through establishment of the prediction model, specific prevention measures can be designed according to risk levels, and the coping capacity and efficiency of the whole medical prevention system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to a method and system for processing and analyzing physical examination data. Background Art

[0002] The technical field of data analysis is to analyze data through mathematical statistical methods and extract useful information from it, involving data collection, processing, evaluation, and interpretation, aiming to help individuals and organizations make data-based decisions. Data analysis can be applied to various different fields, such as finance, market research, healthcare, scientific research, and government management, etc. A variety of tools and technologies are used, including statistical software, data visualization tools, and machine learning algorithms, to discover patterns, trends, and correlations in data, and to find solutions that can be used to optimize processes, improve efficiency, and solve complex problems.

[0003] Among them, the method for processing and analyzing physical examination data is an application specific to the field of medical and health, involving systematic analysis of a batch of health data collected during physical examinations to identify health risks, evaluate the probability of disease development, and formulate prevention strategies. Through the analysis of physical examination data, medical institutions can better understand the health status of patient groups, personalize medical services, and provide more accurate health guidance and disease prevention for patients. The analysis requires the use of advanced data processing technologies and statistical methods to ensure the accuracy and effectiveness of data interpretation.

[0004] Although existing data analysis technologies are widely used in the field of medical and health, when dealing with complex and dynamically changing physical examination data, they are limited by the static nature and generalization ability of the analysis model. Dependent on standardized statistical methods, it is difficult to adapt to the uniqueness and non-linear characteristics of individual health data, resulting in the inability to accurately predict health risks or identify abnormal health events in actual operations. When existing technologies conduct health risk assessments, they lack in-depth mining of the correlation between historical health data and current data, making the assessment of health trends lack personalization and foresight. The limitation lies in that it focuses more on descriptive analysis rather than the generation of predictive or defensive strategies. Therefore, in the rapidly changing field of health monitoring, the application effect is limited. For example, when facing a high-risk patient group, standard analysis methods cannot provide sufficient data support to design effective prevention measures, resulting in a lack of timeliness or accuracy in responding to potential health crises and affecting the overall health management effect. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose a method and system for processing and analyzing physical examination data.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions. A method for processing and analyzing physical examination data includes the following steps:

[0007] S1: Extract individual health indicators from the physical examination dataset, classify and organize the health indicators by type and time, and compare the real-time physical examination data with the historical physical examination data to generate differential health indicator data;

[0008] S2: Perform anomaly detection on the differential health indicator data, identify changes in health indicators in the physical examination data, analyze the density difference of the changes, identify abnormal health events through processing the physical examination data, and generate abnormal health event records;

[0009] S3: Based on the abnormal health event records, identify the periodic patterns and non-periodic mutations of abnormal events in the physical examination data, assign health risk levels to the identified patterns, and generate health risk pattern data;

[0010] S4: According to the health risk pattern data, calculate the frequency and degree of the risk pattern by analyzing the physical examination data, and combine with the individual's historical health records to evaluate the potential health trends and generate risk trend assessment data;

[0011] S5: Utilize the risk trend assessment data to identify and predict health events and risk developments in future time periods, adjust the risk level according to the real-time physical examination data and the prediction time, and generate a health event prediction model;

[0012] S6: Use the health event prediction model to design preventive measures for the risk level and potential health events by processing and analyzing the physical examination data, and generate health risk prediction results.

[0013] As a further solution of the present invention, the differential health indicator data includes the difference in heart rate variability, the difference in cholesterol level, and the difference in hemoglobin concentration. The abnormal health event records include abnormal heart rate events, cholesterol warnings, and abnormal fluctuations in hemoglobin. The health risk pattern data includes seasonal fluctuations in abnormal heart rate, non-periodic changes in cholesterol, and hemoglobin mutation events. The risk trend assessment data includes seasonal risk assessment of cardiovascular diseases, long-term cholesterol trend analysis, and statistical frequency of abnormal fluctuations in hemoglobin. The health event prediction model includes a time model for predicting arrhythmia events, a trend model for predicting abnormal cholesterol levels, and an event model for predicting acute changes in hemoglobin. The health risk prediction results include customized diet plans and adjusted exercise plans.

[0014] As a further solution of the present invention, the steps of extracting individual health indicators from the physical examination dataset, classifying and organizing the health indicators by type and time, and comparing the real-time physical examination data with the historical physical examination data to generate differential health indicator data are specifically as follows:

[0015] S101: Analyze each record item in the physical examination data set based on the physical examination data set, extract the data of basic health indicators including heart rate, blood pressure, and blood sugar, file the indicators according to the individual identifier, and obtain the initialized health classification data set;

[0016] S102: According to the initialized health classification data set, sort each type of health indicator according to the date of the physical examination data, separately organize them into category files for cardiovascular and blood sugar, verify that the health indicator data in each category is arranged in chronological order, and generate the sorted physical examination health data;

[0017] S103: Compare the real-time physical examination data with the historical data through the sorted physical examination health data, calculate the change value for each indicator, and record the change of the indicator to generate the differential health indicator data.

[0018] As a further solution of the present invention, the steps of performing anomaly detection on the differential health indicator data, identifying the changes in the health indicators in the physical examination data, analyzing the density difference of the changes, and identifying abnormal health events through the processing of the physical examination data to generate abnormal health event records are specifically as follows:

[0019] S201: Based on the differential health indicator data, calculate the frequency and amplitude of the change of each health indicator, use the set threshold to judge that the change of the health indicator exceeds the normal range, identify potential abnormal health events, and generate the initialized anomaly marking record;

[0020] S202: Utilize the initialized anomaly marking record, adopt the kernel density estimation algorithm, perform density analysis on the marked health indicators in the physical examination data, evaluate the change density and distribution characteristics, identify the trend and dispersion degree of the change of the health indicators, and generate the density difference analysis result;

[0021] S203: Through the density difference analysis result, compare the historical data and abnormal patterns of the health indicators in the physical examination data, identify health risks and potential health problems, and organize and record the identified abnormal events to generate abnormal health event records.

[0022] As a further solution of the present invention, the formula of the kernel density estimation algorithm is as follows:

[0023]

[0024] Among them, f h (x) is the probability density function, w i is the weight of the data point X i , K h,α,β is the kernel function with adjustment parameters, h is the bandwidth, α and β are the kernel function shape adjustment parameters, X i is a single data point, and n is the total number of data points.

[0025] As a further solution of the present invention, based on the abnormal health event records, the steps of identifying the periodic patterns and aperiodic mutations of abnormal events in the physical examination data, assigning health risk levels to the identified patterns, and generating health risk pattern data are specifically as follows:

[0026] S301: Based on the abnormal health event records, by processing the time series of the physical examination data, analyzing the time points and frequencies of each event, identifying seasonal variations and patterns associated with target activities, and generating a periodic pattern analysis result;

[0027] S302: According to the periodic pattern analysis result, process the aperiodic abnormal events, identify unconventional health changes by comparing the deviation degree of the sudden changes from the existing health data, and generate an aperiodic mutation record;

[0028] S303: Combine the aperiodic mutation record with the periodic pattern analysis result, evaluate each identified health risk pattern, assign a risk level to each pattern according to the potential health impact, and generate health risk pattern data.

[0029] As a further solution of the present invention, the steps of calculating the frequency and degree of the risk pattern, combining the individual's historical health records, evaluating the potential health trend, and generating risk trend assessment data by analyzing the physical examination data according to the health risk pattern data are specifically as follows:

[0030] S401: Based on the health risk pattern data, by statistically matching the frequency data and the health records, calculate the frequency and urgency of the health risk pattern, evaluate the relevance to the individual's historical health records, and generate an urgency analysis result;

[0031] S402: Use the urgency analysis result to analyze the trends in the physical examination data, identify the changing trends of health risks, predict the key health problems, and generate a health trend prediction result;

[0032] S403: According to the health trend prediction result, evaluate the individual's health condition in the future time period, formulate a prevention plan for the risk levels in the physical examination data, and generate risk trend assessment data.

[0033] As a further solution of the present invention, the steps of identifying and predicting health events and risk developments in the future time period by using the risk trend assessment data, adjusting the risk level according to the real-time physical examination data and the prediction time, and generating a health event prediction model are specifically as follows:

[0034] S501: Analyze health events within a future time period based on the risk trend assessment data. By comparing with real-time and historical health data, identify health risks in the physical examination data and generate an analysis result of predicted health events.

[0035] S502: Use the analysis result of predicted health events to predict health events and risk development within a target time period. Refer to the influence of seasonal changes and personal health history to generate a time prediction result of health risks.

[0036] S503: Adopt the time prediction result of health risks to adjust the health risk level. According to real-time physical examination data and predicted health events, set warning signals for each risk level and generate a health event prediction model.

[0037] As a further solution of the present invention, the steps of using the health event prediction model to design preventive measures for risk levels and potential health events by processing and analyzing physical examination data to generate a health risk prediction result are specifically as follows:

[0038] S601: Based on the health event prediction model, process and analyze physical examination data, identify health indicators and prediction results associated with risk levels. By identifying and extracting risk health indicators, generate a risk level classification result.

[0039] S602: Use the risk level classification result to design preventive measures for different risk levels, including adjustments to lifestyle, use of preventive medications, and recommendations for regular examinations, and generate a preventive measure design plan.

[0040] S603: Apply the preventive measure design plan to real-time physical examination data, evaluate the adaptability and expected effects of preventive measures, evaluate the influence and effects of preventive measures on predicted health risks, and generate a health risk prediction result.

[0041] A processing and analysis system for physical examination data, which is used to execute the above-mentioned processing and analysis method for physical examination data. The system includes:

[0042] The data classification module extracts the health indicators of an individual from the physical examination dataset, organizes the data according to the indicator type and examination date, compares the real-time physical examination data with the historical data, and generates a health indicator difference table.

[0043] The difference detection module analyzes each indicator difference according to the health indicator difference table, marks the data beyond the normal change range, identifies the abnormal changes of health indicators, and generates an abnormal health indicator record.

[0044] The anomaly analysis module analyzes the marked anomaly data periodically and aperiodically based on the anomaly health index records, assigns a health risk level to each pattern, and generates a health risk pattern table.

[0045] The risk identification module uses the health risk pattern table to calculate the frequency and degree of the risk patterns, combines the historical health data of the individual, evaluates the health trend in a future time period, and generates risk trend assessment data.

[0046] The prevention strategy module analyzes the real-time physical examination data based on the risk trend assessment data, predicts health events and risks in a future time period, designs health management and prevention measures for multiple risk levels, and generates a health risk prediction result.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0048] In the present invention, through comprehensive analysis of physical examination data, detailed classification of individual health indicators and time series comparison are realized, effectively revealing the differences between real-time data and historical data. By accurately identifying abnormal changes in the data, the ability to identify abnormal health events is enhanced, promoting the early detection of health risks. Through detailed analysis of the periodic patterns and aperiodic mutations of abnormal events, risk factors are further classified into levels, providing a more personalized health risk assessment. By combining the frequency and degree of risk patterns with the historical health records of individuals, the ability to evaluate potential health trends is enhanced, making health management more forward-looking and targeted. Through the establishment of a prediction model, not only future health events are predicted, but also specific prevention measures can be designed according to the risk levels, improving the response ability and efficiency of the entire medical prevention system. Brief Description of the Drawings

[0049] Figure 1 It is a schematic diagram of the working process of the present invention;

[0050] Figure 2 It is a detailed flowchart of S1 of the present invention;

[0051] Figure 3 It is a detailed flowchart of S2 of the present invention;

[0052] Figure 4 It is a detailed flowchart of S3 of the present invention;

[0053] Figure 5 It is a detailed flowchart of S4 of the present invention;

[0054] Figure 6 It is a detailed flowchart of S5 of the present invention;

[0055] Figure 7 It is a detailed flowchart of S6 of the present invention;

[0056] Figure 8 This is the system flowchart of the present invention. Detailed implementation manners

[0057] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0058] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the accompanying drawings. These are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation of the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more unless otherwise specifically defined.

[0059] Embodiment 1

[0060] Please refer to Figure 1 , the present invention provides a technical solution, a method for processing and analyzing physical examination data, including the following steps:

[0061] S1: Extract individual health indicators from the physical examination dataset, classify and organize the health indicators according to type and time, combine historical physical examination data, compare the differences between real-time physical examination data and historical physical examination data, and generate differential health indicator data;

[0062] S2: Perform anomaly detection on the differential health indicator data, identify the changes in the health indicators in the physical examination data, analyze the density differences of the changes, identify abnormal health events through the processing of the physical examination data, and generate abnormal health event records;

[0063] S3: Analyze the time sequence in the abnormal health event records, identify the periodic patterns and non-periodic mutations of the abnormal events in the physical examination data, assign health risk levels to the identified patterns, and generate health risk pattern data;

[0064] S4: According to the health risk pattern data, calculate the frequency and degree of each risk pattern, compare the risk health patterns with the individual's historical health records by analyzing the physical examination data, evaluate the potential health trends, and generate risk trend assessment data;

[0065] S5: Utilize the risk trend assessment data to identify and predict health events and risk developments within a future time period, adjust the risk level based on real-time physical examination data and the predicted time, and generate a health event prediction model.

[0066] S6: Use the health event prediction model to design preventive measures for the risk level and potential health events by processing and analyzing physical examination data, and generate health risk prediction results.

[0067] Differentiated health indicator data includes poor heart rate variability, poor cholesterol level, and poor hemoglobin concentration. Abnormal health event records include abnormal heart rate events, cholesterol warnings, and abnormal hemoglobin fluctuations. Health risk pattern data includes seasonal fluctuations in abnormal heart rate, non-periodic changes in cholesterol, and hemoglobin mutation events. Risk trend assessment data includes seasonal risk assessment of cardiovascular diseases, long-term cholesterol trend analysis, and statistics on the frequency of abnormal hemoglobin fluctuations. The health event prediction model includes a time model for predicting arrhythmia events, a trend model for predicting abnormal cholesterol levels, and an event model for predicting acute changes in hemoglobin. Health risk prediction results include customized diet plans and adjusted exercise plans.

[0068] Please refer to Figure 2 , for the steps of extracting individual health indicators from the physical examination dataset, classifying and organizing the health indicators by type and time, and comparing real-time physical examination data with historical physical examination data to generate differentiated health indicator data, specifically:

[0069] S101: Based on the physical examination dataset, analyze each record item in the physical examination data, extract data on basic health indicators including heart rate, blood pressure, and blood sugar, and file the indicators according to the individual identifier. The execution process for obtaining the initialized health classification dataset is as follows;

[0070] Sub-step S101 filters out the three basic health indicators of heart rate, blood pressure, and blood sugar based on the physical examination dataset. For each physical examination record, extract the key data and associate it with the individual identifier to ensure that each piece of data can be traced back to a specific individual. Data cleaning is involved, such as removing invalid or abnormal values, to ensure the accuracy and usability of the dataset. Through processing, the initialized health classification dataset is obtained, and the formula is:

[0071]

[0072] where, D init is the initialized health classification dataset, n is the number of records, hr i is the heart rate value of the i-th record, bp i is the blood pressure value of the i-th record, and bs i is the blood sugar value of the i-th record.

[0073] S102: According to the initialized health classification dataset, sort each type of health indicator according to the date of the physical examination data, and separately organize them into category files for cardiovascular and blood glucose. Verify that the health indicator data in each category is arranged in chronological order. The execution process for generating the sorted physical examination health data is as follows;

[0074] Sub-step S102 sorts the data according to the physical examination date based on the initialized health classification dataset. Create two category files for cardiovascular data (heart rate and blood pressure) and blood glucose data respectively. Ensure that in each category file, the data is arranged in chronological order, facilitating subsequent trend analysis and comparison. By verifying whether the data in each category is correctly arranged, generate the sorted physical examination health data. The formula used is:

[0075]

[0076] where, H sorted is the sorted physical examination health data, n is the number of records, cv i is the cardiovascular data of the i-th record, and sg i is the blood glucose data of the i-th record.

[0077] S103: Through the sorted physical examination health data, compare the real-time physical examination data with the historical data, calculate the change value for each indicator, and record the change of the indicator. The execution process for generating the differential health indicator data is as follows;

[0078] Sub-step S103 compares the real-time physical examination data with the historical data through the sorted physical examination health data. By calculating the change value of each indicator, such as the degree of change in heart rate, blood pressure, and blood glucose, evaluate the change trend of the health status. For each indicator, record the increase or decrease value compared with the previous record. The data will be used to monitor the health changes of individuals and provide more explicit health trend information to doctors and patients. The formula for generating the differential health indicator data is:

[0079]

[0080] where, C delta is the differential health indicator data, n is the number of records, hr i and hr i-1 are the heart rate values of the i-th and i - 1-th records, bp i and bp i-1 are the blood pressure values of the i-th and i - 1-th records, and bs i and bs i-1 are the blood glucose values of the i-th and i - 1-th records.

[0081] Please refer to Figure 3, the steps for performing anomaly detection on differential health indicator data, identifying changes in health indicators in physical examination data, analyzing the density differences in the changes, and identifying abnormal health events through processing of the physical examination data to generate records of abnormal health events are as follows:

[0082] S201: Based on the differential health indicator data, calculate the frequency and amplitude of changes in each health indicator, use a set threshold to determine if the changes in the health indicators exceed the normal range, identify potential abnormal health events, and the execution process for generating an initial anomaly marking record is as follows;

[0083] The sub-steps of S201 analyze the frequency and amplitude of changes in heart rate, blood pressure, and blood sugar based on the differential health indicator data. By analyzing the changes in the indicators, a pre-set threshold is used to determine which changes exceed the normal range, indicating potential abnormal health events. The determination relies on the standard deviation and average rate of change obtained by comparing with historical data. Mark the abnormal changes and generate an initial anomaly marking record. The formula used is:

[0084]

[0085] where, E init is the initial anomaly marking record, n is the number of health indicators, f i is the change frequency of the i-th health indicator, a i is the change amplitude of the i-th health indicator, and T is the set threshold.

[0086] S202: Using the initial anomaly marking record, adopt the kernel density estimation algorithm to perform density analysis on the marked health indicators in the physical examination data, evaluate the change density and distribution characteristics, identify the trends and dispersion degrees of the changes in the health indicators, and the execution process for generating the density difference analysis result is as follows;

[0087] The sub-steps of S202 use the initial anomaly marking record to perform density analysis on the marked health indicators using the kernel density estimation (KDE) method. The aim is to evaluate the change density and distribution characteristics of the marked health indicators, and determine the central tendency and dispersion degree of the changes in the health indicators by calculating the kernel density function. This helps to identify trends where the changes in the health indicators show anomalies or atypical trends and generate the density difference analysis result.

[0088] The formula for the kernel density estimation algorithm is as follows:

[0089]

[0090] where, f h (x) is the probability density function, w i is the weight of the data point X i and K h,α,βis a kernel function with adjustment parameters, h is the bandwidth, α and β are kernel function shape adjustment parameters, and X i is a single data point, and n is the total number of data points.

[0091] The execution process is as follows:

[0092] Set the bandwidth h to determine the smoothness of the kernel function, introduce two new parameters α and β to adjust the shape of the kernel function to better adapt to the characteristics of the data, and for each data point X i assign the weight w i , adjust its influence in density estimation according to the importance or reliability of the data, calculate the kernel function value of each data point, sum and average all the calculation results to obtain the overall density estimation. Confirm the coefficient value method of the weight w i is to analyze the historical change range and frequency of each health indicator X i , use an exponential decay or growth function to give higher weights to the most recent data, making the weights proportional to the timeliness and importance of the data.

[0093] S203: Through the density difference analysis results, compare the historical data and abnormal patterns of health indicators in the physical examination data, identify health risks and potential health problems, and organize and record the identified abnormal events. The execution process for generating abnormal health event records is as follows;

[0094] The S203 sub-step compares the historical data of health indicators in the physical examination data with the identified abnormal patterns through the density difference analysis results. Through comparison, abnormal events related to existing health risks and potential health problems can be identified. The process includes matching the current abnormal pattern with known disease patterns and evaluating potential health risks. Organize and record the identified abnormal events to generate abnormal health event records. The formula used is:

[0095]

[0096] where A event is the abnormal health event record, n is the number of health indicators, o i is the observed abnormal value of the i-th health indicator, and h i is the historical average value of the health indicator.

[0097] Please refer to Figure 4 , based on the abnormal health event record, identify the periodic patterns and non-periodic mutations of abnormal events in the physical examination data, assign health risk levels to the identified patterns, and the steps for generating health risk pattern data are specifically as follows:

[0098] S301: Based on the abnormal health event records, by processing the time series of physical examination data, analyze the time points and frequencies of each event, identify seasonal variations and patterns associated with target activities, and the execution process of generating the periodic pattern analysis results is as follows;

[0099] The sub-steps of S301 process the time series of physical examination data based on the abnormal health event records, and analyze the time points and frequencies of each event. It includes using time series analysis techniques to identify seasonal variations in the data and patterns related to specific target activities (such as festivals, seasonal changes, etc.). Through statistical analysis methods, such as Fourier transform or seasonal decomposition, determine the periodic characteristics of the changes in health indicators, and generate the periodic pattern analysis results. The formula used is:

[0100]

[0101] where, P cycle is the periodic pattern analysis result, n is the number of events, f i is the frequency of the i-th event, t i is the time point of the i-th event, and φ i is the phase difference of the i-th event.

[0102] S302: According to the periodic pattern analysis result, process the non-periodic abnormal events, identify the unconventional health changes by comparing the deviation degree of the sudden changes from the existing health data, and the execution process of generating the non-periodic mutation record is as follows;

[0103] The sub-steps of S302 focus on processing non-periodic abnormal events according to the periodic pattern analysis result, and use statistical anomaly detection techniques, such as outlier analysis, to identify unconventional health changes by comparing the deviation degree of the sudden changes from the existing health data. Non-periodic events indicate new or isolated health problems not covered by the periodic pattern. By analyzing the mutations, generate the non-periodic mutation record. The formula used is:

[0104]

[0105] where, N sporadic is the non-periodic mutation record, n is the number of events, x i is the health indicator value of the i-th event, and μ is the average value of the health indicator.

[0106] S303: Combine the non-periodic mutation record with the periodic pattern analysis result, evaluate each identified health risk pattern, assign a risk level according to the potential health impact, and the execution process of generating the health risk pattern data is as follows;

[0107] The sub-step of S303 combines the aperiodic mutation records with the results of periodic pattern analysis to comprehensively evaluate each identified health risk pattern. This includes using data clustering and risk assessment techniques to assign a risk level to each pattern based on the potential health impacts of various health risk patterns. By comprehensively considering multiple aspects of the health data, health risk pattern data is generated, using the formula:

[0108]

[0109] Where, R risk is the health risk pattern data, n is the number of health patterns, w i is the weight of the i-th health pattern, and r i is the risk level of the i-th health pattern.

[0110] Please refer to Figure 5 , for the steps of generating risk trend assessment data by analyzing the physical examination data, calculating the frequency and degree of the risk pattern, and combining with the individual's historical health record to evaluate the potential health trend based on the health risk pattern data, specifically:

[0111] S401: Based on the health risk pattern data, through the statistical matching of frequency data and health records, calculate the frequency and urgency of the health risk pattern, evaluate the relevance to the individual's historical health record, and the execution process of generating the urgency analysis result is as follows;

[0112] The sub-step of S401 calculates the occurrence frequency and urgency of each health risk pattern based on the health risk pattern data through the statistical matching of frequency data and the individual's historical health record. It involves quantifying the occurrence times and severity of each health risk pattern, and evaluating the relevance between the pattern and the individual's historical health record to generate the urgency analysis result, using the formula:

[0113]

[0114] Where, U urgency is the urgency analysis result, n is the number of health risk patterns, f i is the occurrence frequency of the i-th health risk pattern, and s i is the urgency score of the i-th health risk pattern.

[0115] S402: Use the urgency analysis result to analyze the trend in the physical examination data, identify the changing trend of health risks, predict key health problems, and the execution process of generating the health trend prediction result is as follows;

[0116] The sub-step of S402 uses the result of the urgency analysis to analyze the trends in the physical examination data and identify the changing trends of health risks. The analysis includes using time series analysis or regression models to predict the future development of key health problems. By comprehensively analyzing the historical and current trends of the physical examination data, it helps to identify health risks in advance, provides a scientific basis for timely intervention and treatment, generates the health trend prediction result, and the formula is:

[0117]

[0118] where T trend is the health trend prediction result, n is the number of variables, β0 is the model intercept, β i is the coefficient of the i-th variable, and x i is the historical data of the i-th health risk factor.

[0119] S403: According to the health trend prediction result, evaluate the individual's health status in the future time period, formulate a prevention plan for the risk level in the physical examination data, and the execution process of generating the risk trend assessment data is as follows;

[0120] The sub-step of S403 evaluates the individual's health status in the future time period according to the health trend prediction result. The evaluation is based on the predicted health trend and the individual's risk level, and a specific prevention plan is designed. During the process, the prediction result is combined with the risk level in the physical examination data to determine the priority and scope of the prevention measures. The data helps to formulate a more targeted health management strategy, improve the efficiency and effectiveness of prevention and treatment, and generate the risk trend assessment data. The formula is:

[0121]

[0122] where R assessment is the risk trend assessment data, n is the number of health events, v i is the risk level of the i-th predicted health event, and p i is the intensity of the prevention plan for the risk level.

[0123] Please refer to Figure 6 , using the risk trend assessment data, identify and predict the health events and risk development in the future time period, adjust the risk level according to the real-time physical examination data and the prediction time, and the steps of generating the health event prediction model are specifically as follows:

[0124] S501: Based on the risk trend assessment data, analyze the health events in the future time period, identify the health risks in the physical examination data by comparing with the real-time and historical health data, and the execution process of generating the predicted health event analysis result is as follows;

[0125] The sub-step of S501 analyzes health events that will occur in a future time period based on risk trend assessment data, including comparing prediction data with real-time and historical health data to identify potential health risks in physical examination data. Through comprehensive analysis, the increased health risks in the future are revealed, providing a basis for further preventive measures, and generating the analysis results of predicted health events. The formula is:

[0126]

[0127] Among them, H forecast is the analysis result of predicted health events, n is the number of health indicators, f i is the predicted value of the i-th health indicator in the future time period, and h i is the value of the corresponding historical health indicator.

[0128] S502: Use the analysis result of predicted health events to predict health events and risk development in the target time period. Referring to the influence of seasonal changes and personal health history, the execution process of generating the time prediction result of health risks is as follows;

[0129] The sub-step of S502 uses the analysis result of predicted health events to predict health events and risk development in the target time period. The process takes into account the influence of seasonal changes and personal health history, uses a statistical model to analyze the time distribution and development trend of health risks, provides a time series prediction of future health risks, helps individuals and medical providers prepare corresponding health management strategies, and generates the time prediction result of health risks. The formula is:

[0130]

[0131] Among them, T risk is the time prediction result of health risks, β0 is the model intercept, n is the number of time units, β i is the health risk change coefficient for the i-th time unit, and t i is the i-th time unit.

[0132] S503: Adopt the time prediction result of health risks to adjust the health risk level. Set warning signals for each risk level according to real-time physical examination data and predicted health events. The execution process of generating the health event prediction model is as follows;

[0133] The S503 sub-step uses the health risk time prediction results to adjust the health risk level. Based on the real-time physical examination data and the predicted health events, warning signals are set for each risk level. This includes analyzing the probabilities and potential impacts of each health event, as well as their associations with the current health status, which not only helps predict future health events but also provides specific warning levels for implementing preventive measures to address potential health threats in advance, generating a health event prediction model. The formula is:

[0134]

[0135] Among them, M health is the health event prediction model, n is the number of health events, α i is the warning signal weight of the i-th health event, and r i is the adjustment coefficient corresponding to the risk level.

[0136] Please refer to Figure 7 , when using the health event prediction model, by processing and analyzing the physical examination data, the steps to design preventive measures for the risk level and potential health events and generate the health risk prediction results are as follows:

[0137] S601: Based on the health event prediction model, process and analyze the physical examination data, identify the health indicators and prediction results associated with the risk level. The execution process of generating the risk level classification result by identifying and extracting the risk health indicators is as follows;

[0138] The S601 sub-step, based on the health event prediction model, identifies the health indicators and prediction results associated with the risk level, deeply analyzes each indicator in the physical examination data, and extracts the key health indicators related to the high-risk level. The indicators include blood pressure, blood sugar, cholesterol, etc. Abnormal values indicate high health risks, generating the risk level classification result. The formula is:

[0139]

[0140] Among them, C risk is the risk level classification result, n is the number of health indicators, x i is the value of the i-th health indicator, and k i is the risk weight coefficient of the i-th health indicator.

[0141] S602: Using the risk level classification result, design preventive measures for different risk levels, including lifestyle adjustments, use of preventive medications, and recommendations for regular check-ups. The execution process of generating the preventive measure design plan is as follows;

[0142] Sub-step S602 utilizes the risk level classification results to design preventive measures for different risk levels, including recommending appropriate lifestyle adjustments, preventive medication use, and regular check-ups for individuals with different risk levels. For example, a low-salt diet and regular blood pressure monitoring are recommended for hypertensive patients, and reducing saturated fat intake and using lipid-lowering medications are recommended for patients with high cholesterol. Based on each individual's specific health condition and the predicted risk level, a preventive measure design plan is generated, using the formula:

[0143]

[0144] where P prevention is the preventive measure design plan, n is the number of risk levels, r i is the i-th risk level, and m i is the intensity of the preventive measure for the risk level.

[0145] S603: Apply the preventive measure design plan to the real-time physical examination data, evaluate the adaptability and expected effect of the preventive measures, evaluate the impact and effect of the preventive measures on the predicted health risks, and the execution process for generating the health risk prediction results is as follows;

[0146] Sub-step S603 applies the preventive measure design plan to the real-time physical examination data, evaluates the adaptability and expected effect of the preventive measures, and evaluates the impact and effect on the predicted health risks by simulating and actually applying the measures. The evaluation considers the actual implementation possibility, acceptance, and expected health improvement effect of the measures, provides a quantitative analysis of the effectiveness of the preventive measures, helps improve future health management strategies, and generates the health risk prediction results, using the formula:

[0147]

[0148] where R health is the health risk prediction result, n is the number of preventive measures, p i is the implementation intensity of the i-th preventive measure, and δ i is the change rate of the health indicator expected to be improved by the preventive measure.

[0149] Please refer to Figure 8 , a processing and analysis system for physical examination data. The processing and analysis system for physical examination data is used to execute the above-mentioned processing and analysis method for physical examination data. The system includes:

[0150] The data classification module extracts the health indicators of individuals from the physical examination dataset, organizes the data according to the indicator type and examination date, compares the real-time physical examination data with the historical data, and generates a health indicator difference table;

[0151] The difference detection module analyzes the difference of each indicator according to the health indicator difference table, identifies the data beyond the normal change range, recognizes the abnormal changes of health indicators, and generates abnormal health indicator records;

[0152] The abnormal analysis module conducts periodic and aperiodic analyses on the marked abnormal data through the abnormal health indicator records, assigns health risk levels to each mode, and generates a health risk mode table;

[0153] The risk identification module uses the health risk mode table to calculate the frequency and degree of risk modes, combines the historical health data of an individual, evaluates the health trend in a future time period, and generates risk trend assessment data;

[0154] The prevention strategy module analyzes the real-time physical examination data based on the risk trend assessment data, predicts health events and risks in a future time period, designs health management and prevention measures for multiple risk levels, and generates health risk prediction results.

[0155] The above are only the preferred embodiments of the present invention, and the present invention is not limited to other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for processing and analyzing physical examination data, characterized in that, It includes the following steps: Extract individual health indicators from the physical examination dataset, classify and sort the health indicators by type and time, and compare the real-time physical examination data with the historical physical examination data to generate differential health indicator data; Perform anomaly detection on the differential health indicator data, identify changes in health indicators in the physical examination data, analyze the density differences in the changes, identify abnormal health events through the processing of the physical examination data, and generate abnormal health event records; Based on the abnormal health event records, identify the periodic patterns and aperiodic mutations of abnormal events in the physical examination data, assign health risk levels to the identified patterns, and generate health risk pattern data; According to the health risk pattern data, calculate the frequency and degree of the risk pattern by analyzing the physical examination data, and combine with the individual's historical health records to evaluate the potential health trends and generate risk trend assessment data; Utilize the risk trend assessment data to identify and predict health events and risk developments within a future time period, adjust the risk level according to the real-time physical examination data and the prediction time, and generate a health event prediction model; Use the health event prediction model to design preventive measures for the risk level and potential health events by processing and analyzing the physical examination data, and generate health risk prediction results.

2. The processing and analysis method of physical examination data according to claim 1, characterized in that The differential health indicator data includes the difference in heart rate variability, the difference in cholesterol level, and the difference in hemoglobin concentration. The abnormal health event records include abnormal heart rate events, cholesterol warnings, and abnormal fluctuations in hemoglobin. The health risk pattern data includes seasonal fluctuations in abnormal heart rate, aperiodic changes in cholesterol, and hemoglobin mutation events. The risk trend assessment data includes seasonal risk assessments of cardiovascular diseases, long-term cholesterol trend analysis, and statistics on the frequency of abnormal fluctuations in hemoglobin. The health event prediction model includes a time model for predicting arrhythmia events, a trend model for predicting abnormal cholesterol levels, and an event model for predicting acute changes in hemoglobin. The health risk prediction results include customized diet plans and adjusted exercise plans.

3. The method for processing and analyzing physical examination data according to claim 1, characterized in that, The steps of extracting individual health indicators from the physical examination dataset, classifying and sorting the health indicators by type and time, and comparing the real-time physical examination data with the historical physical examination data to generate differential health indicator data are specifically as follows: Based on the physical examination dataset, analyze each record item in the physical examination data, extract data on basic health indicators including heart rate, blood pressure, and blood sugar, and file the indicators by individual identifier to obtain an initialized health classification dataset; According to the initialized health classification dataset, sort each type of health indicator by the date of the physical examination data, and separately organize them into category files for cardiovascular and blood sugar, verify that the health indicator data in each category is arranged in chronological order, and generate the sorted physical examination health data; Through the sorted physical examination health data, compare the real-time physical examination data with the historical data, calculate the change value for each indicator, and record the changes in the indicators to generate differential health indicator data.

4. The processing and analysis method of physical examination data according to claim 1, characterized in that The steps for performing anomaly detection on the differential health index data, identifying changes in health indices in the physical examination data, analyzing the density differences of the changes, and identifying abnormal health events through the processing of the physical examination data to generate abnormal health event records are as follows: Based on the differential health index data, calculate the frequency and amplitude of changes in each health index, use the set threshold to determine whether the changes in the health index exceed the normal range, identify potential abnormal health events, and generate an initial anomaly marker record; Using the initial anomaly marker record, adopt the kernel density estimation algorithm to perform density analysis on the marked health indices in the physical examination data, evaluate the change density and distribution characteristics, identify the trends and dispersion degrees of the changes in the health indices, and generate a density difference analysis result; Based on the density difference analysis result, compare the historical data and abnormal patterns of the health indices in the physical examination data, identify health risks and potential health problems, and organize and record the identified abnormal events to generate abnormal health event records.

5. The processing and analysis method of physical examination data according to claim 4, wherein, The formula for the kernel density estimation algorithm is as follows: where f h (x) is the probability density function, w i is the weight of the data point X i , K h,α,β is the kernel function with adjustment parameters, h is the bandwidth, α and β are the kernel function shape adjustment parameters, X i is a single data point, and n is the total number of data points.

6. The method for processing and analyzing physical examination data according to claim 1, wherein The steps for identifying the periodic patterns and non-periodic mutations of abnormal events in the physical examination data based on the abnormal health event records, assigning health risk levels to the identified patterns, and generating health risk pattern data are as follows: Based on the abnormal health event records, analyze the time points and frequencies of each event by processing the time series of the physical examination data, identify seasonal changes and patterns associated with target activities, and generate a periodic pattern analysis result; According to the periodic pattern analysis result, process non-periodic abnormal events, identify unconventional health changes by comparing the deviation degree of sudden changes from the existing health data, and generate a non-periodic mutation record; Combine the non-periodic mutation record with the periodic pattern analysis result, evaluate each identified health risk pattern, and assign a risk level to each pattern according to the potential health impact to generate health risk pattern data.

7. The method for processing and analyzing physical examination data according to claim 1, wherein, The steps for generating risk trend assessment data by analyzing the physical examination data, calculating the frequency and degree of the risk pattern, and evaluating the potential health trend in combination with the individual's historical health record based on the health risk pattern data are as follows: Based on the health risk pattern data, calculate the frequency and urgency of the health risk pattern by statistically matching the frequency data and health records, evaluate the relevance to the individual's historical health record, and generate an urgency analysis result; Use the urgency analysis result to analyze the trends in the physical examination data, identify the changing trends of health risks, and predict key health problems to generate a health trend prediction result; According to the health trend prediction result, evaluate the individual's health status in the future time period, formulate a prevention plan for the risk levels in the physical examination data, and generate risk trend assessment data.

8. The processing and analysis method of physical examination data according to claim 1, wherein The steps for identifying and predicting health events and risk developments in the future time period using the risk trend assessment data, and adjusting the risk level according to the real-time physical examination data and prediction time to generate a health event prediction model are as follows: Based on the risk trend assessment data, analyze health events in a future time period, identify health risks in the physical examination data by comparing with real-time and historical health data, and generate an analysis result of predicted health events; Use the analysis result of predicted health events to predict health events and risk developments in a target time period, and generate a time prediction result of health risks with reference to the influence of seasonal changes and personal health history; Adopt the time prediction result of health risks to adjust the health risk level, set warning signals for each risk level according to real-time physical examination data and predicted health events, and generate a health event prediction model.

9. The method for processing and analyzing physical examination data according to claim 1, wherein The steps of using the health event prediction model to generate a health risk prediction result by processing and analyzing physical examination data and designing preventive measures for risk levels and potential health events are specifically as follows: Based on the health event prediction model, process and analyze physical examination data, identify health indicators and prediction results associated with risk levels, and generate a risk level classification result by identifying and extracting risk health indicators; Use the risk level classification result to design preventive measures for different risk levels, including adjustments to lifestyle, use of preventive medications, and recommendations for regular examinations, and generate a preventive measure design plan; Apply the preventive measure design plan to real-time physical examination data, evaluate the adaptability and expected effects of preventive measures, evaluate the impact and effects of preventive measures on predicted health risks, and generate a health risk prediction result.

10. A processing and analysis system for physical examination data, characterized in that, According to the method for processing and analyzing physical examination data according to any one of claims 1-9, the system includes: The data classification module extracts the health indicators of an individual from the physical examination dataset, sorts the data according to the indicator type and examination date, compares the real-time physical examination data with the historical data, and generates a health indicator difference table; The difference detection module analyzes each indicator difference according to the health indicator difference table, marks the data that exceeds the normal change range, identifies the abnormal changes of health indicators, and generates an abnormal health indicator record; The abnormal analysis module performs periodic and non-periodic analysis on the marked abnormal data through the abnormal health indicator record, assigns a health risk level to each pattern, and generates a health risk pattern table; The risk identification module uses the health risk pattern table to calculate the frequency and degree of risk patterns, combines the individual's historical health data, evaluates the health trend in a future time period, and generates risk trend assessment data; The prevention strategy module analyzes real-time physical examination data based on the risk trend assessment data, predicts health events and risks in a future time period, and designs health management and preventive measures for multiple risk levels to generate a health risk prediction result.

Images