Clinical aging comprehensive assessment method and system and storage medium

Abstract

The invention relates to the technical field of data processing, and discloses a clinical aging comprehensive assessment method and system and a storage medium. The method comprises the following steps: acquiring and processing physiological indexes to obtain physiological index data; performing cognitive function test and compensation based on the physiological index data to obtain cognitive evaluation data; analyzing behavior characteristics according to the physiological indexes and the cognitive evaluation data to obtain activity characteristic data; performing matrix processing on the three types of data to obtain senescence feature data; scoring and recombining the senescence characteristic data to obtain senescence quantitative data; and formulating and optimizing an intervention scheme according to the aging quantitative data. According to the invention, the accuracy of the clinical aging comprehensive assessment method is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a clinical aging comprehensive assessment method, system and storage medium. Background Art

[0002] As the global population ages, research on aging assessment and intervention methods is becoming increasingly important. Existing aging assessment methods mainly include physiological indicator monitoring, cognitive function testing, and behavioral characteristic analysis. In terms of physiological indicator monitoring, basic data such as blood pressure, heart rate, and blood sugar are obtained through wearable devices and medical examinations; in terms of cognitive function testing, standardized scales are used to assess cognitive abilities such as memory, attention, and reaction speed; in terms of behavioral characteristic analysis, questionnaires and activity records are used to assess social activities and daily living abilities. These assessment methods provide basic data support for aging research.

[0003] However, the existing technology has the following shortcomings: first, the various evaluation indicators are independent of each other and lack systematic integrated analysis, which leads to one-sided evaluation results and cannot fully reflect the individual's aging status; second, the impact of human physiological rhythms is not considered in the evaluation process, and the test results are easily affected by time factors; third, the data collection and processing methods are relatively simple and fail to effectively deal with data noise and outliers, affecting the accuracy of the evaluation results; finally, there is a lack of optimization mechanism for personalized intervention plans based on evaluation results, making it difficult to provide accurate health management recommendations for the elderly. Summary of the invention

[0004] The present application provides a clinical aging comprehensive assessment method, system and storage medium for establishing a multi-dimensional and dynamic clinical aging comprehensive assessment method, realizing the systematic integration of physiological indicators, cognitive functions and behavioral characteristics, and providing accurate aging assessment results and personalized intervention plans through data mining and machine learning technologies.

[0005] In a first aspect, the present application provides a clinical aging comprehensive assessment method, which comprises: collecting data on physiological indicators, and obtaining physiological indicator data through data denoising and normalization processing; based on the physiological indicator data, performing time period testing on cognitive function, performing compensation processing according to the rhythm curve, and obtaining cognitive assessment data through data fusion analysis; tracking and recording behavioral characteristics based on the physiological indicator data and the cognitive assessment data, and obtaining activity characteristic data through social frequency calculation and network structure analysis; performing matrix construction and data decomposition on the physiological indicator data, the cognitive assessment data and the activity characteristic data, and obtaining aging characteristic data through threshold calculation and data reconstruction; performing score calculation and feature reorganization on the aging characteristic data, and obtaining aging quantitative data through baseline comparison and weight adjustment; performing target decomposition on the intervention plan based on the aging quantitative data, and obtaining the intervention optimization plan through effect evaluation and parameter adjustment.

[0006] In a second aspect, the present application provides a clinical aging comprehensive assessment system, the clinical aging comprehensive assessment system comprising:

[0007] The acquisition module is used to collect data of physiological indicators and obtain physiological indicator data through data denoising and normalization processing;

[0008] A compensation module, used to perform a time period test on cognitive function based on the physiological index data, perform compensation processing according to the rhythm curve, and obtain cognitive evaluation data through data fusion analysis;

[0009] An analysis module, used to track and record behavioral characteristics based on the physiological indicator data and the cognitive assessment data, and obtain activity characteristic data through social frequency calculation and network structure analysis;

[0010] A decomposition module, for performing matrix construction and data decomposition on the physiological index data, the cognitive assessment data and the activity characteristic data, and obtaining aging characteristic data through threshold calculation and data reconstruction;

[0011] A reorganization module is used to perform score calculation and feature reorganization on the aging feature data, and obtain aging quantitative data through baseline comparison and weight adjustment;

[0012] The adjustment module is used to decompose the intervention plan according to the quantitative data of aging, and obtain the intervention optimization plan through effect evaluation and parameter adjustment.

[0013] A third aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned clinical aging comprehensive assessment method.

[0014] In the technical solution provided by the present application, interference factors in physiological indicator data are effectively eliminated through data denoising and normalization processing, thereby improving the quality and reliability of original data; cognitive function period testing based on rhythmic curves overcomes the defect of traditional testing methods that ignore the influence of human physiological rhythms, and realizes a more scientific cognitive ability assessment; tracking and recording of behavioral characteristics and network structure analysis break through the limitation of traditional evaluation methods that only focus on a single behavioral indicator, and realize a comprehensive grasp of social activities and behavioral patterns; the application of matrix construction and data decomposition technology solves the problem of difficulty in unified processing of multi-source heterogeneous data, and realizes the effective integration of physiological, cognitive and behavioral data; the process of score calculation and feature recombination overcomes the shortcomings of traditional scoring methods that are highly subjective, and establishes objective and quantitative evaluation standards; the formulation of intervention plans based on quantitative aging data realizes a seamless connection from assessment to intervention, and improves the accuracy of comprehensive clinical aging assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0016] Figure 1 This is a schematic diagram of an embodiment of the comprehensive clinical aging assessment method in the embodiments of the present application;

[0017] Figure 2 A schematic diagram of a rhythm curve in an embodiment of the present application;

[0018] Figure 3 Schematic diagram of an embodiment of a comprehensive clinical aging assessment system in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The embodiments of the present application provide a method, system and storage medium for comprehensive assessment of clinical aging. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0020] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , one embodiment of the clinical aging comprehensive assessment method in the embodiments of the present application includes:

[0021] Step S101, collecting data of physiological indicators, and obtaining physiological indicator data through data denoising and normalization processing;

[0022] Step S102: Based on the physiological index data, the cognitive function is tested during a period of time, compensation processing is performed according to the rhythm curve, and cognitive assessment data is obtained through data fusion analysis;

[0023] Step S103: tracking and recording the behavior characteristics based on the physiological index data and cognitive assessment data, and obtaining the activity characteristic data through social frequency calculation and network structure analysis;

[0024] Step S104: Matrix construction and data decomposition are performed on the physiological index data, cognitive assessment data and activity characteristic data, and aging characteristic data is obtained through threshold calculation and data reconstruction;

[0025] Step S105, score calculation and feature reorganization are performed on the aging feature data, and aging quantitative data is obtained through baseline comparison and weight adjustment;

[0026] Step S106: Decompose the intervention plan into targets according to the quantitative data of aging, and obtain the intervention optimization plan through effect evaluation and parameter adjustment.

[0027] It is understandable that the execution subject of the present application may be a clinical aging comprehensive assessment system, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0028] Specifically, the data of physiological indicators are collected through a multi-channel physiological sensor array, and the collected indicators include basic physiological data such as blood pressure, heart rate, blood sugar and cholesterol levels. Data collection adopts two methods: real-time monitoring and regular testing. Real-time monitoring continuously records dynamic indicators such as heart rate and blood pressure through wearable devices such as smart watches; regular testing obtains static indicators such as blood sugar and cholesterol through medical examinations. The collected raw data is denoised by wavelet transform to eliminate random noise and baseline drift generated during the measurement process, and then the indicators of different dimensions are mapped to a unified interval [0,1] through maximum and minimum value normalization to form standardized physiological indicator data. Based on the standardized physiological indicator data, the individual is tested for cognitive function in different time periods in combination with the circadian rhythm. The test content covers multiple dimensions such as memory, attention, and reaction speed. The test period is divided into three time windows: morning, noon and evening. During the test, the test results are dynamically compensated by the physiological rhythm curve to eliminate the influence of time factors on cognitive performance. Specifically, 7-9 am is the cognitive function active period, and the test results need to be multiplied by a compensation coefficient of 0.9; 12-14 pm is the cognitive function trough period, and the test results need to be multiplied by a compensation coefficient of 1.2; 17-19 pm is the cognitive function secondary active period, and the test results need to be multiplied by a compensation coefficient of 1.0. Through multi-level data fusion analysis, the compensated test data is integrated into unified cognitive assessment data.

[0029] On the basis of obtaining physiological index data and cognitive assessment data, the behavioral characteristics of individuals are tracked and recorded around the clock. The activity trajectory, social media usage, and exercise data of individuals are recorded through smart terminals. In the social frequency calculation stage, indicators such as the number of daily social activities, the duration of a single social interaction, and the number of social objects are counted. The network structure analysis quantitatively describes the breadth and depth of the social network by constructing a social relationship network graph and calculating topological features such as node degree centrality and betweenness centrality. Through these analyses, the activity feature data is finally obtained. For the three types of data obtained: physiological index data, cognitive assessment data, and activity feature data, a third-order tensor matrix is ​​first constructed to uniformly represent the features of different dimensions. The feature tensor is decomposed by high-order singular value decomposition to extract the main feature components. In the threshold calculation stage, a density-based clustering method is used to determine the feature screening threshold and eliminate redundant and noise features. Finally, the screened features are reconstructed by non-negative matrix decomposition to obtain aging feature data reflecting the individual's aging status.

[0030] Quantitative scoring calculations were performed on aging feature data. First, the feature distribution law was analyzed through probability density estimation, and a scoring standard based on age stratification was established. In the feature reorganization process, principal component analysis was used to reconstruct the feature space and extract the most representative feature combination. In the baseline comparison phase, the individual score was compared with the baseline of the same age group, and the confidence interval of the score was estimated by Bootstrap sampling. The entropy weight method was used for weight adjustment, and the weight coefficient was dynamically allocated according to the amount of information of the feature, and finally the aging quantitative data was obtained. According to the aging quantitative data, a personalized intervention plan was formulated, and the intervention goals were decomposed through the hierarchical analysis method to establish an intervention system including exercise prescription, nutritional conditioning, cognitive training and other dimensions. Grey correlation analysis was used in the effect evaluation stage to calculate the correlation between intervention measures and effect indicators. In the parameter adjustment process, the intervention parameters were optimized by genetic algorithm, such as exercise intensity, training duration, nutritional supplement dosage, etc., and finally a dynamically optimized intervention plan was formed.

[0031] For example, a medical institution conducted a three-month aging assessment and intervention on a 65-year-old man. During the physiological data collection phase, the average blood pressure of the elderly was recorded as 135 / 85mmHg, resting heart rate was 72 beats / min, fasting blood glucose was 5.8mmol / L, and total cholesterol was 5.2mmol / L. Cognitive function tests showed that his cognitive scores at different times were: 85 points in the morning (76.5 points after compensation), 65 points at noon (78 points after compensation), and 75 points in the evening (75 points after compensation). Behavior tracking records showed that he participated in social activities 4-5 times a week, each lasting an average of 2 hours, and his fixed social circle included 8 people. Through data processing and analysis, it was finally concluded that the elderly's aging characteristic score was 3.2 points (out of 5 points), and he was in a mild aging state. Based on this assessment result, an intervention plan was formulated, including 3 times a week of moderate-intensity aerobic exercise (30 minutes each time), cognitive game training (1 hour per day), and social activity arrangements (adding 1 group activity per week). After three months of intervention, their aging characteristic scores dropped to 2.8 points, indicating that the intervention was effective.

[0032] In the embodiments of the present application, through data denoising and normalization processing, the interference factors in the physiological indicator data are effectively eliminated, and the quality and reliability of the original data are improved; the cognitive function period test based on the rhythm curve overcomes the defect of the traditional test method ignoring the influence of the human physiological rhythm, and realizes a more scientific cognitive ability assessment; the tracking and recording of behavioral characteristics and the network structure analysis break through the limitation of the traditional evaluation method that only focuses on a single behavioral indicator, and realizes a comprehensive grasp of social activities and behavioral patterns; the application of matrix construction and data decomposition technology solves the problem that multi-source heterogeneous data is difficult to process in a unified manner, and realizes the effective integration of physiological, cognitive and behavioral data; the process of score calculation and feature recombination overcomes the shortcomings of the traditional scoring method with strong subjectivity, and establishes an objective and quantitative evaluation standard; the formulation of intervention plans based on aging quantitative data realizes a seamless connection from evaluation to intervention, and improves the accuracy of comprehensive clinical aging evaluation.

[0033] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0034] (1) A multi-channel physiological sensor array collects heart rate, blood pressure, and blood glucose in different time windows, and obtains the original physiological data set after data pre-screening and time series completion;

[0035] (2) Based on the spectral characteristics of the original physiological data set, adaptive wavelet decomposition is used to separate high-frequency noise and baseline drift, and empirical mode decomposition is combined to eliminate artifacts to obtain a denoised data set;

[0036] (3) Extract periodic fluctuation characteristics from the denoised data set, combine weighted moving average and nonlinear smoothing technology to enhance the stability of the data, remove trend terms through piecewise regression, and obtain stable data;

[0037] (4) Construct a multi-dimensional feature space for the stabilized data, perform data dimension reduction and reconstruction through principal component rotation and maximum information entropy criterion, and combine Z-score standardization to obtain standardized feature data;

[0038] (5) Setting an automatically adjusted dynamic threshold boundary for the standardized feature data, identifying outliers through density clustering and anomaly detection algorithms, and correcting the data in combination with expert rules to obtain corrected data;

[0039] (6) The corrected data are grouped through hierarchical clustering, and key features are extracted according to the information gain criterion. After feature combination and data alignment, physiological indicator data are obtained.

[0040] Specifically, in the process of collecting and processing physiological index data, the multi-channel physiological sensor array first collects data. The sensor array includes a blood pressure sensor, a heart rate sensor and a blood glucose sensor, and uses different time windows for hierarchical collection. Blood pressure data is collected every 30 minutes, heart rate data is collected every 5 minutes, and blood glucose data is collected every 2 hours. In the data pre-screening stage, validity tests are used to eliminate obvious abnormal values, such as data with heart rate exceeding the range of 40-120 times / minute. For missing data in the collection process, linear interpolation is used to fill the time series to ensure the continuity of the data, thereby obtaining the original physiological data set. After obtaining the original physiological data set, noise processing is required. First, the data is spectrally analyzed to determine the main frequency components of the signal. Adaptive wavelet decomposition technology is used to select appropriate wavelet basis functions to perform multi-scale decomposition on the data. For heart rate data, db4 wavelet is selected to perform 5-layer decomposition to extract signal components in different frequency bands respectively. High-frequency noise is mainly concentrated in the wavelet coefficients of the 1st and 2nd layers, and is processed by soft threshold denoising. The baseline drift is mainly reflected in the 4th and 5th layer wavelet coefficients, which are corrected by median filtering. For the artifacts that still exist after processing, the empirical mode decomposition technology is used for further optimization. The empirical mode decomposition decomposes the signal into several intrinsic mode functions, and by analyzing the frequency characteristics of each intrinsic mode function, the artifact components are identified and removed, and finally the noise reduction data set is obtained.

[0041] When performing stability processing on the denoised data set, the periodic fluctuation characteristics of the data are first extracted. The periodic components of the signal are analyzed by Fourier transform to identify the main physiological rhythm characteristics. The weighted moving average technology is then used to preliminarily smooth the data, and the window size is adjusted according to the sampling frequency of different indicators, such as a 15-minute window for heart rate data and a 1-hour window for blood pressure data. On this basis, nonlinear smoothing technology is introduced, and the local regression algorithm is used to further process the data to effectively retain the local characteristics of the data. The long-term trend items of the data are identified by segmented regression analysis and removed from the original data to obtain stabilized data. When constructing the feature space for stabilized data, each physiological indicator is represented as a different dimension. The feature space is transformed by the principal component rotation technology to find the optimal feature representation direction. The maximum information entropy criterion is used for feature selection to retain the feature combination with the largest amount of information. The selected features are Z-score standardized to convert indicators of different dimensions into a unified scale space to obtain standardized feature data.

[0042] After obtaining the standardized feature data, outlier detection and processing are required. Set the dynamic threshold boundary, and the threshold is automatically adjusted according to the data distribution characteristics. Use the DBSCAN density clustering algorithm to cluster the data and identify outliers in the data. Combined with the normal range rules of physiological indicators formulated by experts, the identified outliers are corrected. For outliers that cannot be determined, the local mean is replaced to obtain corrected data. Finally, feature extraction and organization are performed on the corrected data. A hierarchical clustering algorithm is used to group the data, and a hierarchical tree structure is constructed based on the similarity between the data. The importance of different features is evaluated by the information gain criterion, and the most valuable feature combination for aging assessment is selected. The selected features are aligned in time series to obtain the final physiological indicator data.

[0043] For example, a medical institution collects and processes physiological data of a 70-year-old man. During the 24-hour continuous monitoring, the heart rate sensor collects data every 5 minutes, with a total of 288 data points, and the original heart rate range is between 55-95 beats / minute. After removing the measurement noise through adaptive wavelet decomposition, the heart rate data range is narrowed to 58-88 beats / minute. After weighted moving average processing (15-minute window), a smoother heart rate change curve is obtained. Blood pressure data is collected every 30 minutes, with a total of 48 data points, and the original systolic blood pressure range is between 115-165mmHg. After noise reduction and smoothing, the blood pressure data range is adjusted to 125-155mmHg. Blood sugar data is collected every 2 hours, with a total of 12 data points, and the value range is between 4.8-7.2mmol / L. Through feature space construction and standardization, these three groups of indicators are unified into the standard normal distribution space. Density cluster analysis found two abnormal value points (heart rate 92 beats / minute, blood pressure 162mmHg), which were corrected by expert rules to obtain more reasonable values ​​(heart rate 85 beats / minute, blood pressure 150mmHg). Through feature extraction and combination, a complete data set reflecting the changes in the elderly's 24-hour physiological state was formed.

[0044] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0045] (1) Associating and pairing physiological indicator data with timestamps, extracting physiological fluctuation patterns through circadian rhythm analysis algorithms, and obtaining rhythm baseline data;

[0046] (2) Establish a multi-dimensional cognitive test matrix for the rhythmic baseline data, obtain test sequence data by grading the task difficulty and dividing the time window, and combining the preset rhythmic curve and attention fluctuation curve;

[0047] (3) Setting dynamic response thresholds for test sequence data, quantifying test results through reaction time statistics and error pattern analysis, and obtaining raw cognitive data;

[0048] (4) Extracting time series features from the original cognitive data, separating cognitive fluctuation patterns by wavelet packet decomposition, and evaluating complexity by entropy calculation to obtain cognitive feature data;

[0049] (5) Perform multi-scale analysis on cognitive feature data, extract instantaneous phase information through Hilbert transform, and perform data correction in combination with rhythm compensation algorithm to obtain compensated data;

[0050] (6) The compensated data is processed through multi-level data fusion, and feature integration is performed in combination with information entropy weight allocation. Multi-dimensional features are adaptively combined to obtain cognitive assessment data.

[0051] Specifically, the physiological indicator data are associated and paired with time stamps to establish a time series database. The 24-hour change pattern of physiological indicators is analyzed by the circadian rhythm analysis algorithm, including key time periods such as the early morning temperature rise period (06:00-08:00), the midday temperature stability period (12:00-14:00) and the night temperature drop period (20:00-22:00). The rhythmic baseline data reflects the periodic change characteristics of human physiological indicators and provides a time reference for subsequent cognitive tests. After obtaining the rhythmic baseline data, a multi-dimensional cognitive test matrix is ​​established, including three dimensions: memory test, attention test and reaction speed test. The task difficulty is divided into three levels: elementary, intermediate and advanced. Elementary tasks include simple digital memory (3-5 digits) and graphic recognition (2-3 targets); intermediate tasks include medium-complexity text memory (5-7 words) and multi-target tracking (4-6 targets); advanced tasks include complex logical reasoning, multi-task coordination and other contents. The time window division takes into account the rhythmic curve of the human body, such as Figure 2 As shown in the figure, the curve shows that the human body's cognitive ability shows obvious fluctuations throughout the day: 8:00-10:00 in the morning is the first peak period of cognitive ability, with high excitability of the cerebral cortex, which is suitable for complex cognitive tasks; 12:00-14:00 in the afternoon is the trough period of cognitive ability, when the brain needs to rest; 15:00-17:00 in the afternoon is the second peak period of cognitive ability, which is suitable for cognitive tasks of medium difficulty. The attention fluctuation curve shows the changing pattern of human attention level: the attention level gradually rises in the morning, reaching a peak at 9:00-11:00, which lasts for about 2 hours; there is a short drop in attention at 14:00-15:00 in the afternoon; and the second attention peak occurs at 18:00-20:00 in the evening. Combining the characteristics of these two curves, the test tasks are arranged in the most suitable time window to form scientific test sequence data.

[0052] A dynamic response threshold is set for the test sequence data, and the threshold is dynamically adjusted with the test difficulty and time window. The reaction time statistics adopt a dual recording method, recording both the first reaction time and the total time to complete the task. Error pattern analysis includes three types: omission errors (missing the target stimulus), incorrect response (responding to non-target stimuli) and slow response (exceeding the normal reaction time range). Weighted calculation is performed on each error type to obtain the original cognitive data. The time series features are extracted from the original cognitive data through wavelet packet decomposition, and the cognitive fluctuation pattern is decomposed into components in different frequency bands. Wavelet packet decomposition can accurately capture subtle changes in cognitive performance and distinguish short-term fluctuations from long-term trends. By calculating the entropy value of each frequency band, the complexity and stability of cognitive performance are evaluated, and finally cognitive feature data are obtained.

[0053] The cognitive feature data is analyzed at multiple scales, and the instantaneous phase information of the data is extracted using Hilbert transform. Hilbert transform can accurately reflect the dynamic change characteristics of cognitive performance and provide a basis for rhythm compensation. The rhythm compensation algorithm corrects the test results according to the fluctuations of cognitive ability in different periods of time to obtain compensated data. Finally, the compensated data is subjected to multi-level fusion processing. The information entropy weight allocation scientifically allocates weight coefficients according to the amount of information of each indicator. The multi-dimensional features are adaptively combined to form the final cognitive assessment data.

[0054] For example, a 68-year-old man was evaluated for cognitive function. When the first test was conducted at 8:30 in the morning, the average simple reaction time was 285 milliseconds and the accuracy was 92%. Considering that cognitive ability is rising at this time, the rhythm curve compensation coefficient is 0.9, and the corrected reaction time is 256.5 milliseconds. In the memory test, the accuracy of the immediate recall of the 6-digit sequence was 83%, and the accuracy of the delayed recall (after 30 minutes) dropped to 75%. The test at 13:00 noon showed that the simple reaction time was extended to 320 milliseconds and the accuracy dropped to 85%. At this time, the rhythm compensation coefficient was 1.2, and the corrected reaction time was 384 milliseconds. In the test at 16:00 in the afternoon, the reaction time recovered to 290 milliseconds and the accuracy increased to 90%. Through wavelet packet decomposition, it was found that the main frequency of its cognitive fluctuations was concentrated in the range of 0.1-0.3Hz, indicating that cognitive function is relatively stable. Information entropy analysis shows that the weight of the reaction time index is 0.4, the weight of the memory index is 0.35, and the weight of the attention index is 0.25. The comprehensive calculation showed that the elderly person's cognitive function score was 78 points (out of 100 points), which is at the middle level of people of the same age. This score fully takes into account the fluctuations in cognitive performance throughout the day and eliminates the influence of time factors through rhythm compensation.

[0055] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0056] (1) Time-align the physiological indicator data and cognitive assessment data, construct a behavioral time series benchmark through multi-source data association analysis, and obtain benchmark feature data;

[0057] (2) Setting multi-dimensional behavior monitoring indicators for baseline feature data, classifying and counting daily activities through behavior pattern recognition algorithms, and obtaining behavioral statistical data;

[0058] (3) Extracting social interaction information from behavioral statistics, quantifying the interaction intensity and frequency through social network topology analysis, and obtaining social feature data;

[0059] (4) Expand the social feature data in time and space, divide the social group structure through spectral clustering algorithm, extract features by combining centrality calculation, and obtain network structure data;

[0060] (5) Use graph theory to analyze network structure data for social relationship mining, and combine it with community evolution characteristics for dynamic tracking to obtain social dynamic data;

[0061] (6) Multi-dimensional feature combination is performed on social dynamic data, and activity feature data is obtained through time series correlation analysis and feature fusion processing combined with the evolution law of behavioral patterns.

[0062] Specifically, the physiological indicator data and cognitive assessment data are aligned according to the timestamp. The physiological indicator data includes continuous measurement values ​​such as heart rate and blood pressure, and the cognitive assessment data includes cognitive test results of each time period. Through time series analysis technology, data with different sampling frequencies are unified to the same time scale to build a unified behavioral time series benchmark. The multi-source data association analysis adopts the sliding time window method, with the window size set to 30 minutes and each sliding for 15 minutes. The correlation of various indicators in the window is analyzed to obtain the benchmark feature data. Based on the benchmark feature data, a multi-dimensional behavior monitoring indicator system is set, including basic indicators such as daily activity intensity, activity duration, and activity frequency. The behavioral pattern recognition algorithm first classifies the collected behavioral data and divides daily activities into four levels: resting activities (such as watching TV and reading), light activities (such as walking and doing housework), moderate activities (such as brisk walking and Tai Chi), and heavy activities (such as running and playing ball). The duration and frequency of each type of activity are counted to generate behavioral statistical data.

[0063] Social interaction information is extracted from behavioral statistics, including face-to-face conversations, participation in group activities, telephone communications and other forms. Social network topology analysis first constructs a social relationship graph, in which nodes represent individuals and edges represent social connections. The intensity of social interaction is calculated by recording the duration and number of participants of each social activity. The frequency of social interaction is obtained by counting the number of social activities per unit time. Combining these indicators, social feature data is obtained. The social feature data is expanded in time and space dimensions to analyze the temporal distribution law and spatial distribution characteristics of social activities. The spectral clustering algorithm constructs the adjacency matrix of the social network, calculates the eigenvalues ​​and eigenvectors of the Laplace matrix, and naturally classifies social groups. Centrality calculation includes degree centrality (the number of direct social connections), betweenness centrality (the degree of serving as a social bridge) and closeness centrality (the average social distance with others), which together constitute the network structure data.

[0064] The network structure data is deeply mined through graph theory analysis. First, the basic characteristics of the network are calculated, such as the average path length, clustering coefficient, etc. The community evolution characteristics are obtained by tracking the changes in social networks at different time points, including new social connections, changes in connection strength, and other information to form social dynamic data. Finally, the social dynamic data is feature fused to combine time series features, network topology features, and behavior pattern features. Time series correlation analysis evaluates the changing trends of various features over time, and the evolution laws of behavior patterns reflect the long-term changes in social behavior. The final activity feature data is obtained through weighted feature fusion.

[0065] For example, a behavioral characteristic analysis of a 72-year-old man was conducted for one month. First, the heart rate data (once every 5 minutes) and the cognitive test results (3 times a day) were time-aligned, and it was found that the heart rate increased slightly during social activities (an average increase of 8-10 times / minute), and the cognitive performance was relatively stable. Behavioral pattern recognition showed that resting activities accounted for 45% (about 6.5 hours), light activities accounted for 35% (about 5 hours), moderate activities accounted for 15% (about 2 hours), and heavy activities accounted for 5% (about 0.5 hours). Social network analysis showed that the elderly had a fixed social circle of 12 people, of which 5 were core social circles. The average daily social time was 3.2 hours, including morning exercise group activities (1.5 hours), neighborhood conversations (1 hour) and family interactions (0.7 hours). Spectral clustering divided his social network into three main groups: sports partners, neighbors and family. Centrality calculation showed that the betweenness centrality of the sports partner group was high, indicating that it played an important bridge role in group communication. One month of dynamic tracking found that his social activities showed regularity, participating in 2-3 group activities per week, and the intensity and frequency of social activities remained stable. Through feature fusion analysis, the elderly's social activity score was 85 points (out of 100 points), showing good social participation and interaction quality.

[0066] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0067] (1) Physiological index data, cognitive assessment data, and activity feature data are assembled into a multidimensional matrix using a tensor construction algorithm, and data alignment is performed based on feature mapping relationships to obtain feature tensor data;

[0068] (2) Performing high-order singular value decomposition on the feature tensor data, screening the principal components by the nuclear norm minimization criterion, and obtaining the decomposed feature data;

[0069] (3) Construct a covariance matrix for the decomposed feature data, quantify the feature correlation through subspace analysis, and obtain correlation data;

[0070] (4) Extract key feature combinations from the correlation data, reconstruct the feature structure through non-negative matrix decomposition, and optimize it in combination with sparse constraints to obtain reconstructed data;

[0071] (5) Setting multi-level dynamic thresholds for the reconstructed data, performing feature classification through sorting statistics and hierarchical clustering, and obtaining hierarchical feature data;

[0072] (6) The graded feature data are compared and analyzed with the group baseline, and the features are combined and optimized through weighted fusion processing to obtain aging feature data.

[0073] Specifically, three types of data (physiological indicator data, cognitive assessment data, and activity feature data) are organized into a three-dimensional tensor structure through a tensor construction algorithm. The three dimensions of the tensor correspond to individuals, feature types, and time series, respectively. The feature mapping relationship establishes the correspondence between different types of data, and maps data with different sampling frequencies to a unified time axis through time tags to form structured feature tensor data. The feature tensor data is subjected to high-order singular value decomposition to decompose the multidimensional data into core tensors and factor matrices. By setting the nuclear norm minimization criterion, the most important principal components for data characterization are screened out, noise and redundant information are removed, and streamlined decomposed feature data are obtained.

[0074] Construct a covariance matrix for the decomposed feature data and calculate the correlation coefficients between different features. Subspace analysis finds the main feature combination direction through eigenvalue decomposition, quantifies the feature correlation, and obtains correlation data. When extracting key feature combinations from correlation data, non-negative matrix decomposition method is used for feature reconstruction. Given the original data matrix X∈R m×n (m is the number of samples, n is the number of features), we need to find two non-negative matrices W∈R m×k and H∈R k×n (k is the number of potential features), such that:

[0075] X≈WH+E

[0076] Among them, W represents the basis matrix, H represents the coefficient matrix, and E is the error term. By introducing the sparse constraint of the L1 norm:

[0077] min||X-WH|| 2 F +λ||W||1+μ||H||1

[0078] Among them, ||·|| F represents the Frobenius norm, ||·||1 represents the L1 norm, and λ and μ are regularization parameters. In this way, reconstructed data is obtained. Multi-level dynamic thresholds are set for the reconstructed data, and the threshold boundaries are determined according to the data distribution characteristics. The dividing points of each level are determined by sorting statistics, and the features are hierarchically grouped by combining the hierarchical clustering algorithm to obtain hierarchical feature data. The hierarchical feature data are compared and analyzed with the pre-established group baseline data. By weighted fusion, the feature combination is optimized and integrated in combination with the importance of different features, and finally the aging feature data is obtained.

[0079] For example, a group of elderly people (75 people) in a medical institution were analyzed for aging characteristics. First, the data of each person's physiological indicators (15 items), cognitive assessment (8 items) and activity characteristics (10 items) were constructed into a three-dimensional tensor of 75×33×24 (24 represents the time point recorded in hours). After high-order singular value decomposition, the principal components with energy accounting for more than 90% were selected, and the data dimension was reduced to 75×20×24. Correlation analysis found that the correlation coefficient between blood pressure and cognitive reaction speed was -0.72, indicating that cognitive function decreased slightly when blood pressure increased. Through non-negative matrix decomposition, the 33-dimensional original features were reconstructed into 12 key feature combinations. The set dynamic threshold divided the samples into three levels: mild aging (25 people), moderate aging (35 people) and severe aging (15 people). Finally, through weighted fusion (physiological indicator weight 0.4, cognitive assessment weight 0.35, activity feature weight 0.25), the aging characteristic score of each individual was obtained.

[0080] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0081] (1) The aging feature data are stratified according to the distribution of different age groups, the feature distribution law is analyzed through probability density estimation, and the discrete features are processed into continuous ones based on the principles of mathematical statistics to obtain the feature distribution data;

[0082] (2) Calculate the standard score of each dimension indicator in the feature distribution data, quantify the importance of the indicator through entropy weight analysis, initialize and configure the weight vector in combination with multi-objective programming theory, and obtain the weight coefficient data;

[0083] (3) Construct a multi-level scoring system based on the weight coefficient data, score the feature combination through fuzzy comprehensive evaluation, and calibrate the scoring rules based on the expert knowledge base to obtain the scoring matrix data;

[0084] (4) Extract key evaluation dimensions from the rating matrix data, reconstruct the feature space through principal component analysis, calculate the feature similarity based on the distance metric, and obtain feature reorganization data;

[0085] (5) Compare and analyze the feature recombinant data with the population baseline, estimate the confidence interval through Bootstrap sampling, and correct the outliers by combining Bayesian inference to obtain baseline comparison data;

[0086] (6) The baseline comparison data are optimized and combined through a dynamic programming algorithm, and the weight coefficients are dynamically adjusted in combination with a feedback compensation mechanism. Through multi-dimensional feature fusion and normalization processing, quantitative aging data are obtained.

[0087] Specifically, the aging feature data are stratified. According to age, the data are divided into four levels: youth (18-44 years old), middle-aged (45-59 years old), young old age (60-74 years old) and old age (over 75 years old). The distribution law of the characteristics in each age group is analyzed by the kernel density estimation method, and the discrete characteristics (such as social frequency) are processed continuously using the spline interpolation method to obtain continuous feature distribution data. The standard score conversion is performed on each indicator in the feature distribution data, and the indicators of different dimensions are unified into the standard normal distribution interval. Entropy weight analysis evaluates the contribution of each indicator to the overall evaluation by calculating its information entropy. Combined with the multi-objective programming theory, considering the mutual influence between indicators, the initial value of the weight is set, and the weight coefficient data is generated.

[0088] A multi-level scoring system is constructed based on the weight coefficient data, and the fuzzy comprehensive evaluation method is used to quantitatively score the feature combination. In the fuzzy comprehensive evaluation, the evaluation object set U = u1, u2, ..., u n , the evaluation index set V = v1, v2, ..., v m , comment set E = e1, e2, ..., e k , then the fuzzy evaluation matrix R = (r ij ) n×k , where r ij It represents the membership of the i-th evaluation object to the j-th comment. The evaluation result B is calculated by the following formula:

[0089]

[0090] Where A is the weight vector (α1, α2, ..., α n ), represents the fuzzy composition operator, b k Indicates the final evaluation result of the kth comment. The scoring rules are adjusted in combination with the expert knowledge base to finally obtain the scoring matrix data.

[0091] The key evaluation dimensions are extracted from the score matrix data through principal component analysis, and the feature space is reconstructed. The similarity between features is calculated through various distance measurement methods such as Euclidean distance and Manhattan distance to form feature reorganization data. The feature reorganization data is compared with the group baseline. The Bootstrap sampling method is used to generate statistical distribution by repeated sampling 1000 times, and the 95% confidence interval is calculated. For outliers outside the confidence interval, Bayesian inference is used to correct them to obtain baseline comparison data.

[0092] Finally, the baseline comparison data is optimized and combined through dynamic programming algorithm. The feedback compensation mechanism dynamically adjusts the weight coefficient according to the scoring results, and maps the final score to the range of 0-100 through normalization to obtain aging quantitative data.

[0093] For example, a medical research project conducted a quantitative assessment of aging on 500 subjects of different ages. First, they were stratified by age: 150 people in the 18-44 age group, 150 people in the 45-59 age group, 120 people in the 60-74 age group, and 80 people in the 75-year-old and older group. Through kernel density estimation analysis, it was found that physiological indicators (such as blood pressure and heart rate) showed different distribution characteristics in different age groups: the standard deviation of the young group was smaller (blood pressure standard deviation ±5mmHg), while the dispersion of the elderly group was larger (blood pressure standard deviation ±12mmHg). The standard score calculation unified each indicator to a distribution with a mean of 0 and a standard deviation of 1. Entropy weight analysis showed that the weight of physiological indicators was 0.4, the weight of cognitive function was 0.35, and the weight of social activities was 0.25. In the fuzzy comprehensive evaluation, the score levels were divided into five levels: excellent (90-100 points), good (80-89 points), general (70-79 points), poor (60-69 points), and poor (below 60 points). The principal component analysis retained the first five principal components that explained 85% of the total variance. The normal range determined by Bootstrap sampling was: blood pressure index 90-110, cognitive function index 85-115, social activity index 80-120. Finally, through feature fusion, it was concluded that the average score of the 18-44 age group was 88.5 points, the average score of the 45-59 age group was 82.3 points, the average score of the 60-74 age group was 76.8 points, and the average score of the group over 75 years old was 70.2 points.

[0094] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0095] (1) The quantitative data of aging are prioritized by indicators of each dimension through hierarchical analysis, and the intervention target system is constructed in combination with the target decomposition theory. The correlation between targets is quantified through coupling analysis to obtain the target decomposition data;

[0096] (2) Perform multi-objective dynamic programming on the target decomposition data, calculate the allocation of intervention resources through a convex optimization algorithm, and screen the strategy space in combination with constraint conditions to obtain initial intervention data;

[0097] (3) Construct a feedback evaluation matrix based on the initial intervention data, dynamically evaluate the intervention effect through grey correlation analysis, and analyze the effect trend by combining time series feature extraction to obtain effect evaluation data;

[0098] (4) Extract key performance indicators from the effect evaluation data, quantify the intervention efficiency through data envelopment analysis, verify the significance of the effect through difference test, and obtain the efficacy analysis data;

[0099] (5) Perform multidimensional parameter sensitivity analysis on the performance analysis data, optimize the parameter combination through genetic algorithm, explore the parameter space with local search strategy, and obtain parameter optimization data;

[0100] (6) The parameter optimization data is optimized through the reinforcement learning algorithm, the intervention plan is dynamically adjusted in combination with the adaptive control theory, and the plan is comprehensively evaluated through multi-criteria decision-making to obtain the intervention optimization plan.

[0101] Specifically, the quantitative data of aging were analyzed hierarchically, and the evaluation indicators were divided into three dimensions: physiological level (including basic indicators such as blood pressure, heart rate, and blood sugar), cognitive level (including cognitive functions such as memory, attention, and reaction speed), and social activity level (including social frequency, activity intensity, etc.). The priority of each indicator was determined through hierarchical analysis: the physiological level accounted for 40%, the cognitive level accounted for 35%, and the social activity level accounted for 25%. The goal decomposition theory decomposes the overall intervention goal into specific executable sub-goals, such as decomposing the blood pressure control goal into detailed goals such as diet adjustment, exercise planning, and medication guidance. The coupling degree analysis calculates the degree of mutual influence between different goals, such as the positive correlation between exercise intervention and cognitive function improvement, and finally forms the goal decomposition data. The goal decomposition data was optimized using a multi-objective dynamic programming method, considering multiple resource constraints such as time, manpower, and material resources. The convex optimization algorithm optimally allocated the intervention resources, such as weekly exercise time allocation and cognitive training frequency arrangement. The constraints include individual physical fitness limitations, time investment upper limit, and accessibility of medical resources. Linear programming was used to screen out feasible strategy combinations that meet all constraints to form initial intervention data.

[0102] A feedback evaluation matrix was constructed for the initial intervention data, including short-term effect indicators (such as improvement in immediate response ability) and long-term effect indicators (such as stability of physiological indicators). Grey correlation analysis evaluated the effectiveness of various intervention measures by calculating the correlation between intervention measures and effect indicators. Time series feature extraction analyzed the temporal evolution of intervention effects, including immediate effects, cumulative effects, and sustainability, to obtain effect evaluation data. Key performance indicators were extracted from the effect evaluation data, including dimensions such as improvement magnitude, improvement speed, and stability. Data envelopment analysis calculated the input-output ratio of each intervention measure and evaluated the efficiency of the intervention. The statistical significance of the intervention effect was verified through difference tests (such as paired t-tests) to comprehensively form efficacy analysis data.

[0103] A multidimensional parameter sensitivity analysis is performed on the performance analysis data to examine the influence of various intervention parameters (such as exercise intensity, training duration, intervention frequency, etc.) on the effect. The genetic algorithm optimizes the parameter combination through operations such as crossover and mutation, and the local search strategy conducts refined exploration near the optimal solution to obtain parameter optimization data. Finally, the parameter optimization data is input into the reinforcement learning algorithm to optimize the intervention strategy through repeated experiments and reward mechanisms. Adaptive control theory dynamically adjusts the intervention parameters based on individual feedback, and multi-criteria decision-making comprehensively considers multiple dimensions such as effect, cost, and sustainability to form the final intervention optimization plan.

[0104] For example, a medical institution conducted a three-month intervention optimization for a 68-year-old man. Hierarchical analysis showed that the elderly needed to focus on cognitive function (weight increased to 45%), followed by physiological indicators (35%), and social activities (20%). Target decomposition breaks down cognitive function improvement into specific tasks such as attention training (twice a day, 20 minutes each time) and memory training (three times a week, 30 minutes each time). Multi-objective planning takes into account the constraint that the elderly can invest no more than 2 hours a day, and arranges a gradient training plan: the first month is mainly low-intensity cognitive training (total duration 60 minutes / day), the second month adds social activities (45 minutes of cognitive training + 45 minutes of social activities / day), and the third month adds moderate exercise (40 minutes of cognitive training + 40 minutes of social activities + 30 minutes of exercise / day). Grey correlation analysis shows that the correlation between cognitive training and attention improvement is 0.85, with the highest correlation. Efficacy analysis shows that after three months, the attention test score increased by 25%, memory increased by 20%, and the frequency of social activities increased by 35%. Parameter optimization found that 9-11 a.m. is the best time for cognitive training, the training duration is 20-25 minutes, and the frequency is maintained twice a day for the best effect. Based on these findings, reinforcement learning has developed a personalized long-term intervention plan, including fixed training time, progressive difficulty settings and flexible activity combinations.

[0105] The above describes the clinical aging comprehensive evaluation method in the embodiment of the present application. The following describes the clinical aging comprehensive evaluation system in the embodiment of the present application. Figure 3 In the embodiment of the present application, one embodiment of the clinical aging comprehensive assessment system includes:

[0106] The acquisition module is used to collect data of physiological indicators and obtain physiological indicator data through data denoising and normalization processing;

[0107] A compensation module, used to perform a time period test on cognitive function based on the physiological index data, perform compensation processing according to the rhythm curve, and obtain cognitive evaluation data through data fusion analysis;

[0108] An analysis module, used to track and record behavioral characteristics based on the physiological indicator data and the cognitive assessment data, and obtain activity characteristic data through social frequency calculation and network structure analysis;

[0109] A decomposition module, for performing matrix construction and data decomposition on the physiological index data, the cognitive assessment data and the activity characteristic data, and obtaining aging characteristic data through threshold calculation and data reconstruction;

[0110] A reorganization module is used to perform score calculation and feature reorganization on the aging feature data, and obtain aging quantitative data through baseline comparison and weight adjustment;

[0111] The adjustment module is used to decompose the intervention plan according to the quantitative data of aging, and obtain the intervention optimization plan through effect evaluation and parameter adjustment.

[0112] Through the coordinated cooperation of the above components, data denoising and normalization processing are used to effectively eliminate interference factors in physiological indicator data, thereby improving the quality and reliability of original data. The cognitive function period test based on the rhythm curve overcomes the defect of traditional testing methods that ignore the influence of human physiological rhythms, and realizes a more scientific cognitive ability assessment. The tracking and recording of behavioral characteristics and network structure analysis break through the limitation of traditional evaluation methods that only focus on a single behavioral indicator, and realize a comprehensive grasp of social activities and behavioral patterns. The application of matrix construction and data decomposition technology solves the problem of difficulty in unified processing of multi-source heterogeneous data, and realizes the effective integration of physiological, cognitive and behavioral data. The process of score calculation and feature reorganization overcomes the shortcomings of traditional scoring methods that are highly subjective, and establishes objective and quantitative evaluation standards. The formulation of intervention plans based on quantitative aging data realizes a seamless connection from assessment to intervention, and improves the accuracy of comprehensive clinical aging assessment.

[0113] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein instructions are stored in the computer, and when the instructions are executed on a computer, the computer executes the steps of the clinical aging comprehensive assessment method.

[0114] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A clinical comprehensive aging assessment method, characterized in that: The clinical comprehensive aging assessment method comprises: Collect data of physiological indicators, and obtain physiological indicator data through data denoising and normalization; Based on the physiological index data, the cognitive function is tested during a period of time, compensation processing is performed according to the rhythm curve, and cognitive assessment data is obtained through data fusion analysis; According to the physiological indicator data and the cognitive assessment data, the behavioral characteristics are tracked and recorded, and the activity characteristic data are obtained through social frequency calculation and network structure analysis; For the physiological index data, the cognitive assessment data and the activity characteristic data, matrix construction and data decomposition are performed, and aging characteristic data are obtained through threshold calculation and data reconstruction; Performing score calculation and feature reorganization on the aging feature data, and obtaining aging quantitative data through baseline comparison and weight adjustment; According to the quantitative data of aging, the intervention plan is decomposed into targets, and the intervention optimization plan is obtained through effect evaluation and parameter adjustment.

2. The clinical aging comprehensive assessment method according to claim 1, characterized in that: The physiological index data is collected and the physiological index data is obtained by data denoising and normalization, including: The multi-channel physiological sensor array collects heart rate, blood pressure, and blood sugar in different time windows, and obtains the original physiological data set after data pre-screening and time series completion; According to the spectral characteristics of the original physiological data set, adaptive wavelet decomposition is used to separate high-frequency noise and baseline drift, and empirical mode decomposition is combined to eliminate artifacts to obtain a denoised data set; Extracting periodic fluctuation characteristics from the denoised data set, enhancing the stability of the data by combining weighted moving average and nonlinear smoothing techniques, removing trend items by piecewise regression, and obtaining stabilized data; Constructing a multi-dimensional feature space for the stabilized data, performing data dimension reduction and reconstruction through principal component rotation and maximum information entropy criteria, and combining Z-score standardization to obtain standardized feature data; Setting an automatically adjusted dynamic threshold boundary for the standardized feature data, identifying outliers through density clustering and anomaly detection algorithms, and correcting the data in combination with expert rules to obtain corrected data; The corrected data are grouped by hierarchical clustering, and key features are extracted according to the information gain criterion. After feature combination and data alignment, physiological index data are obtained.

3. The clinical aging comprehensive assessment method according to claim 1, characterized in that: Based on the physiological index data, the cognitive function is tested during a period of time, compensation processing is performed according to the rhythm curve, and cognitive assessment data is obtained through data fusion analysis, including: The physiological index data is associated and paired with the timestamp, and the physiological fluctuation law is extracted through the circadian rhythm analysis algorithm to obtain the rhythm baseline data; A multi-dimensional cognitive test matrix is ​​established for the rhythmic benchmark data, and test sequence data is obtained by grading the task difficulty and dividing the time window, combining the preset rhythmic curve and attention fluctuation curve; A dynamic response threshold is set for the test sequence data, and the test results are quantified through reaction time statistics and error pattern analysis to obtain original cognitive data; Extracting time series features from the original cognitive data, separating cognitive fluctuation patterns by wavelet packet decomposition, and evaluating complexity by entropy calculation to obtain cognitive feature data; Performing multi-scale analysis on the cognitive feature data, extracting instantaneous phase information through Hilbert transform, and performing data correction in combination with a rhythm compensation algorithm to obtain compensated data; The compensated data is processed through multi-level data fusion, feature integration is performed in combination with information entropy weight allocation, and multi-dimensional features are adaptively combined to obtain cognitive assessment data.

4. The clinical aging comprehensive assessment method according to claim 1, characterized in that: The behavior characteristics are tracked and recorded based on the physiological indicator data and the cognitive assessment data, and activity characteristic data are obtained through social frequency calculation and network structure analysis, including: Time-aligning the physiological indicator data and the cognitive assessment data, constructing a behavioral time series benchmark through multi-source data association analysis, and obtaining benchmark feature data; Setting multi-dimensional behavior monitoring indicators for the baseline feature data, classifying and counting daily activities through a behavior pattern recognition algorithm, and obtaining behavior statistical data; Extracting social interaction information from the behavioral statistical data, and quantitatively calculating the interaction intensity and frequency through social network topology analysis to obtain social feature data; Expanding the social feature data in time and space dimensions, dividing the social group structure by using a spectral clustering algorithm, extracting features in combination with centrality calculation, and obtaining network structure data; The network structure data is analyzed through graph theory to mine social relationships, and dynamically tracked in combination with community evolution characteristics to obtain social dynamic data; The social dynamic data is subjected to multi-dimensional feature combination, and activity feature data is obtained through time series correlation analysis and feature fusion processing combined with the evolution law of behavior patterns.

5. The clinical aging comprehensive assessment method according to claim 1, characterized in that: The step of constructing a matrix and decomposing data for the physiological index data, the cognitive assessment data, and the activity characteristic data, and obtaining aging characteristic data through threshold calculation and data reconstruction includes: The physiological index data, the cognitive assessment data and the activity feature data are assembled into a multi-dimensional matrix by a tensor construction algorithm, and data are aligned in combination with a feature mapping relationship to obtain feature tensor data; Performing high-order singular value decomposition on the characteristic tensor data, screening the principal components by a nuclear norm minimization criterion, and obtaining decomposed characteristic data; Constructing a covariance matrix for the decomposed feature data, and performing quantitative calculation on feature correlation through subspace analysis to obtain correlation data; Extract key feature combinations from the correlation data, reconstruct the feature structure through non-negative matrix decomposition, and optimize it in combination with sparse constraints to obtain reconstructed data; Setting multi-level dynamic thresholds for the reconstructed data, performing feature classification through sorting statistics and hierarchical clustering, and obtaining classified feature data; The hierarchical feature data are compared and analyzed with the group baseline, and the features are combined and optimized through weighted fusion processing to obtain aging feature data.

6. The clinical comprehensive aging assessment method according to claim 1, characterized in that: The aging characteristic data is scored and reorganized, and the aging quantitative data is obtained through baseline comparison and weight adjustment, including: The aging characteristic data are stratified according to the distribution of different age groups, the characteristic distribution law is analyzed by probability density estimation, and the discrete characteristics are processed continuously in combination with the principle of mathematical statistics to obtain characteristic distribution data; Calculate the standard score of each dimension indicator in the characteristic distribution data, quantify the importance of the indicator through entropy weight analysis, initialize and configure the weight vector in combination with multi-objective programming theory, and obtain weight coefficient data; A multi-level scoring system is constructed for the weight coefficient data, the feature combination is scored through fuzzy comprehensive evaluation, and the scoring rules are corrected in combination with the expert knowledge base to obtain the scoring matrix data; Extract key evaluation dimensions from the rating matrix data, reconstruct the feature space through principal component analysis, calculate feature similarity in combination with a distance metric criterion, and obtain feature reorganization data; Comparative analysis is performed on the characteristic recombinant data and the group baseline, the confidence interval is estimated by Bootstrap sampling, and the outliers are corrected by combining Bayesian inference to obtain baseline comparison data; The baseline comparison data is subjected to feature optimization combination through a dynamic programming algorithm, and the weight coefficient is dynamically adjusted in combination with a feedback compensation mechanism. Through multi-dimensional feature fusion and normalization processing, aging quantitative data is obtained.

7. The clinical aging comprehensive assessment method according to claim 1, characterized in that: The intervention program is decomposed into targets according to the aging quantitative data, and an intervention optimization program is obtained through effect evaluation and parameter adjustment, including: The quantitative data of aging are prioritized by indicators of each dimension through hierarchical analysis, an intervention target system is constructed in combination with the target decomposition theory, and the correlation between targets is quantified through coupling degree analysis to obtain target decomposition data; Perform multi-objective dynamic programming on the target decomposition data, allocate and calculate the intervention resources through a convex optimization algorithm, and screen the strategy space in combination with constraint conditions to obtain initial intervention data; Constructing a feedback evaluation matrix for the initial intervention data, dynamically evaluating the intervention effect through grey correlation analysis, analyzing the effect trend in combination with time series feature extraction, and obtaining effect evaluation data; Extract key performance indicators from the effect evaluation data, quantify the intervention efficiency through data envelopment analysis, verify the significance of the effect through difference test, and obtain efficacy analysis data; Perform multi-dimensional parameter sensitivity analysis on the performance analysis data, optimize the parameter combination through genetic algorithm, explore the parameter space in combination with local search strategy, and obtain parameter optimization data; The parameter optimization data is optimized through a reinforcement learning algorithm, the intervention plan is dynamically adjusted in combination with adaptive control theory, and the plan is comprehensively evaluated through multi-criteria decision-making to obtain an intervention optimization plan.

8. A clinical aging comprehensive assessment system, used to implement the clinical aging comprehensive assessment method according to any one of claims 1 to 7, characterized in that: The clinical aging comprehensive assessment system comprises: The acquisition module is used to collect data of physiological indicators and obtain physiological indicator data through data denoising and normalization processing; A compensation module, used to perform a time period test on cognitive function based on the physiological index data, perform compensation processing according to the rhythm curve, and obtain cognitive evaluation data through data fusion analysis; An analysis module, used to track and record behavioral characteristics based on the physiological indicator data and the cognitive assessment data, and obtain activity characteristic data through social frequency calculation and network structure analysis; A decomposition module, for performing matrix construction and data decomposition on the physiological index data, the cognitive assessment data and the activity characteristic data, and obtaining aging characteristic data through threshold calculation and data reconstruction; A reorganization module is used to perform score calculation and feature reorganization on the aging feature data, and obtain aging quantitative data through baseline comparison and weight adjustment; The adjustment module is used to decompose the intervention plan according to the quantitative data of aging, and obtain the intervention optimization plan through effect evaluation and parameter adjustment.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the clinical comprehensive aging assessment method according to any one of claims 1 to 7 is implemented.

Images