Cardiovascular and cerebrovascular disease risk assessment management system based on health data accurate driving

Abstract

The invention relates to the technical field of health monitoring, in particular to a cardiovascular and cerebrovascular disease risk assessment management system based on accurate driving of health data, and the system comprises a rhythm sampling tracking module, a fluctuation mode joint recognition module, an abnormal joint change labeling module, a priority follow-up sorting module and a risk location division module. According to the method, the difference between systolic pressure and diastolic pressure extreme values in a continuous period is dynamically compared, the heart rate fluctuation stability screening condition is combined, blood oxygen saturation and heart rate variability data are collected in a linkage mode, the sampling frequency is improved, and a multi-parameter combination structure with direction consistency and time coincidence is constructed; and further extracting change fragments with consistent trends and generating a difference sequence, performing quantitative comparison in combination with body temperature amplitude variation characteristics, realizing accurate positioning of a high-collaboration section through time axis aggregation analysis, and effectively enhancing the recognition capability of risk fluctuation fragments in a short period and the continuous tracking effect of a dynamic process.

Description

Technical Field

[0001] The present invention relates to the field of health monitoring technology, and in particular to a cardiovascular and cerebrovascular disease risk assessment and management system based on precise health data driving. Background Art

[0002] The field of health monitoring technology involves tracking and assessing individual health status through the continuous collection, recording, and analysis of human physiological parameters and vital signs. Core issues include multi-parameter physiological data acquisition, sensor signal conversion, electronic health record construction, risk factor analysis, and early warning models. Its overall technical system encompasses wearable device data interfaces, remote data transmission and processing, medical parameter feature extraction, and association rule mining. It is an interdisciplinary field integrating knowledge from multiple disciplines, including medicine, information, and communications, and is widely used in scenarios such as chronic disease management, elderly care, and public health intervention. Traditional cardiovascular and cerebrovascular disease risk assessment and management systems, driven by precise health data, use historical health data, physiological parameters, lifestyle habits, and previous disease information as the basis for static assessments by building a specific rule base. These systems employ statistical risk factor scoring models or linear regression risk calculations, combining single or combined parameters such as blood pressure, heart rate, and blood lipids to quantitatively score individual cardiovascular and cerebrovascular disease risk. These systems typically rely on existing epidemiological research findings to establish risk stratification criteria. They primarily collect data through electronic questionnaires, calculate risk levels based on risk score tables, and use database queries and comparisons to compare high-risk population characteristics to complete risk assessment and management decisions.

[0003] The existing technology uses a static evaluation method based on a rule base, which only covers fixed parameter comparison models and lacks a response mechanism for real-time fluctuations. It is difficult to form feature linkage judgments in the short-term physiological parameter offset stage. When multiple parameters produce low-amplitude non-continuous changes, it cannot trigger an effective evaluation process. Especially in the stage where blood pressure and heart rate fluctuate slightly, a single indicator is not enough to meet the risk score triggering conditions, resulting in the omission of key changes, which can easily cause heart and brain-related abnormalities to expand from weak linkage changes to obvious sudden states, and cannot continuously monitor and identify dynamic collaborative trends in advance. Summary of the Invention

[0004] The purpose of the present invention is to address the shortcomings of the existing technology and propose a cardiovascular and cerebrovascular disease risk assessment and management system based on precise health data.

[0005] To achieve the above objectives, the present invention adopts the following technical solutions: A cardiovascular and cerebrovascular disease risk assessment and management system based on precise health data driving includes: The rhythm sampling and tracking module collects continuous systolic and diastolic blood pressure, calculates the difference between the two segments, determines the direction and amplitude of changes, and synchronously calls blood oxygen and heart rate variability when the heart rate is stable and increases the sampling frequency. After tracking three segments, it determines whether there is regression and generates a rhythm response state. The fluctuation pattern recognition module calls the carotid artery delay, blood oxygen gradient, and heart rate variation according to the rhythm response state, determines whether the change directions of the three are consistent, selects combinations with consistent directions and overlapping time sequences, and generates a heart-brain synchronization fluctuation combination group; The abnormal joint variation annotation module extracts the heart rate variability and blood oxygen data of the heart-brain synchronous fluctuation combination group, determines whether they show a consistent trend, and if so, marks the time and combination, classifies and records them, and generates a joint variation highlight segment; The priority follow-up sorting module extracts the heart rate difference sequence of the joint change highlight segment, determines whether the trend is strengthening, combines the body temperature change amplitude, analyzes the direction of indicator change, and generates a response order arrangement; The risk location division module reads the first response segment in the response order arrangement, cross-compares the combined group segments, and determines whether they are concentratedly distributed. If so, it is marked as a priority observation target to generate a cardiovascular and cerebrovascular disease risk assessment result.

[0006] As a further solution of the present invention, the rhythm response state includes the systolic and diastolic pressure fluctuation amplitude offset, the heart rate fluctuation stability mark, the blood oxygen and heart rate variability linkage characteristics, and the sampling frequency control mark. The heart-brain synchronous fluctuation combination group includes the consistency of the carotid artery pulsation delay direction, the blood oxygen gradient amplitude coordination, and the heart rate instantaneous change time series coincidence. The joint change highlight segment includes the heart rate variability change trend, the blood oxygen saturation change trend, the joint change trend start and end time, and the combination association label. The response order arrangement includes the heart rate difference numerical sequence, the difference enhancement change trend, the body temperature amplitude comparison index, and the consistency of the change direction of the three physiological parameters. The cardiovascular and cerebrovascular disease risk assessment results include the time series overlap density, the fluctuation concentrated distribution segment, and the priority observation target labeling information.

[0007] As a further solution of the present invention, the combination with consistent direction and overlapping time sequence refers to a combination of physiological indicators with the same changing trend and synchronous changing time points within the same time period; The indicator change direction refers to the trend direction of the physiological parameter showing continuous increase, decrease and stability over a period of time.

[0008] As a further solution of the present invention, the rhythm sampling and tracking module includes: The pressure difference comparison submodule obtains the systolic and diastolic pressure values ​​of consecutive cycles, extracts the maximum and minimum differences between the two segments, subtracts them, and records the offset direction and change amplitude to generate a pressure difference offset sequence; The heart rate fluctuation screening submodule calls the pressure difference offset sequence to detect whether the heart rate fluctuation in the sampling period is in a stable range. If so, the blood oxygen and heart rate variability are collected to obtain stable feature data; The response state determination submodule increases the sampling frequency according to the stable characteristic data, tracks whether the three-stage pressure difference offset changes return to the original range, calculates and obtains the joint fluctuation trend of the three-stage pressure difference offset value and the change rate, and obtains the rhythm response state.

[0009] As a further solution of the present invention, the wave pattern association module includes: The indicator extraction submodule obtains the carotid artery pulsation delay, blood oxygen gradient amplitude, and heart rate instantaneous change values ​​at the corresponding time point based on the marked sampling stage in the rhythm response state, uniformly numbers the three indicators according to the sampling time sequence, establishes a cross-indicator time axis mapping relationship, and obtains the original sequence of the three fluctuations; The direction consistency judgment submodule calls the three original fluctuation sequences, compares the fluctuation directions of the three indicators within the same sampling period, and determines whether they are all rising, falling, or stable. Data points with inconsistent directions are eliminated, and only time periods with consistent directions are retained to obtain the same-direction fluctuation interval; The synchronous combination screening submodule identifies whether there is overlap in the sampling points of the indicators on the time axis according to the marked time periods in the same-direction fluctuation interval, screens indicator combinations with consistent directions and synchronized sampling points, and generates a heart-brain synchronous fluctuation combination group.

[0010] As a further solution of the present invention, the abnormal joint change marking module includes: The variation index extraction submodule extracts the heart rate variability and blood oxygen saturation data in the corresponding sampling stage based on the combinations in the heart-brain synchronization fluctuation combination group, uniformly numbers the time periods, establishes a mapping relationship between the indicators and the sampling segments, and generates a variation index time series; The trend consistency judgment submodule calls the variation index time series, judges the change direction of heart rate variability and blood oxygen saturation value in the continuous sampling segment, selects the continuous stages with the same fluctuation direction, establishes the start and end time indexes, and obtains the trend consistency interval information; The highlighted segment construction submodule integrates the combination content, start and end time periods, and grouping according to the trend consistent interval information, classifies and labels the stage behaviors that meet the conditions, and records them into the database to obtain the joint change highlighted segments.

[0011] As a further solution of the present invention, the priority follow-up sorting module includes: The difference sequence construction submodule extracts the maximum and minimum heart rate values ​​based on the joint change highlighted segments, calculates the difference between the two, constructs the difference records corresponding to each segment, and then arranges them in sampling time sequence to form a complete sequence to obtain the heart rate difference sequence; The temperature fluctuation comparison submodule calls the heart rate difference sequence, synchronously extracts the segmented body temperature variation amplitude, compares whether the body temperature variation amplitude shows a synchronous enhancement trend with the current heart rate difference amplitude, calculates the joint offset trend degree of each segment of heart rate difference and body temperature fluctuation, selects sampling segments with consistent trends, and generates a trend synchronization segment index; The multi-parameter direction analysis submodule synchronizes the segment index according to the trend, analyzes the change direction of the heart rate, blood pressure, and blood oxygen in the segment, and determines whether the three are rising, falling, and stable at the same time. If they are consistent, the combined features are recorded and the response sorting level of the current segment is marked to obtain the response order arrangement.

[0012] As a further solution of the present invention, the risk location classification module includes: The first segment reading submodule extracts the time information corresponding to the first response segment based on the response order arrangement, verifies the start and end time and index position, and constructs a mapping relationship between the segment and the entire time series to obtain the time range of the first response; The overlap segment judgment submodule calls the first response time range and cross-matches it with the time periods of all combinations in the heart-brain synchronization fluctuation combination group to screen whether there are multiple combinations that repeatedly appear within the time range covered by the response segment. If the combinations highly overlap in the target time period, they are identified as overlap concentration intervals. The centralized segment labeling submodule performs priority labeling according to the time series information and combination quantity distribution of the overlapping centralized intervals, and simultaneously records the combination type, coverage and distribution characteristics to obtain the cardiovascular and cerebrovascular disease risk assessment results.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by dynamically comparing the differences between the extreme values ​​of systolic and diastolic blood pressure in continuous cycles, combined with the screening conditions for heart rate fluctuation stability, the blood oxygen saturation and heart rate variability data are jointly collected and the sampling frequency is increased, a multi-parameter combination structure with directional consistency and time overlap is constructed, and the change segments with consistent trends are further extracted and a difference sequence is generated. The quantitative comparison is combined with the body temperature variation characteristics, and the precise positioning of the highly coordinated segments is achieved through time axis aggregation analysis, which effectively enhances the ability to identify risk fluctuation segments in short cycles and the continuous tracking effect of dynamic processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a system flow chart of the present invention; Figure 2 This is a flow chart of the rhythm sampling and tracking module of the present invention; Figure 3 This is a flow chart of the wave pattern recognition module of the present invention; Figure 4 This is a flow chart of the abnormal linkage marking module of the present invention; Figure 5 This is a flow chart of the priority follow-up sorting module of the present invention; Figure 6 This is a flow chart of the risk location division module of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, 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 intended to limit the present invention.

[0016] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0017] See also Figure 1 The cardiovascular and cerebrovascular disease risk assessment and management system based on precise health data includes: The rhythm sampling and tracking module obtains the systolic and diastolic blood pressure values ​​of continuous cycles, subtracts the maximum and minimum differences between two consecutive segments, compares the difference offset direction and change amplitude, and then detects whether the heart rate fluctuation during the period is within a stable range. If so, it synchronously calls the blood oxygen saturation and heart rate variability data during the collection period and simultaneously increases the sampling frequency. After tracking three segments, it determines whether the amplitude has returned to the original fluctuation range and generates a rhythm response state. The fluctuation pattern recognition module uses the marked sampling phase in the rhythm response state to call the carotid artery pulse delay, blood oxygen gradient amplitude, and heart rate instantaneous change values ​​at the corresponding time point to determine whether the change direction of these three indicators is consistent within the time period. It then compares the synchronized time points and selects combinations with consistent directions and overlapping time sequences to generate a heart-brain synchronized fluctuation combination group. The abnormal joint variation annotation module extracts heart rate variability and blood oxygen saturation data during the sampling phase based on the combinations in the heart-brain synchronization fluctuation combination group. It determines whether the two show consistent changes in trend during the continuous acquisition phase. If so, it marks the start and end time and combination content, records the phases, and classifies and stores the change behavior to generate joint variation highlight segments. The priority follow-up sorting module extracts the maximum and minimum heart rate values ​​in the highlighted segment and calculates the difference to form a difference sequence. It then determines whether the difference trend is continuously increasing. It also extracts the temperature variation of the segment and performs a quantitative comparison to determine whether the changes in heart rate, blood pressure, and blood oxygen in the segment are consistent in direction. It analyzes the fluctuation information and generates a response order arrangement. The risk location division module reads the content of the first response segment in the response order arrangement, cross-compares the overlapping segments in the time series with the heart and brain synchronous fluctuation combination group, and determines whether the repeated fluctuations are concentrated in the same time axis range. If there is a concentrated segment, it is marked as a priority observation target to generate a cardiovascular and cerebrovascular disease risk assessment result.

[0018] The rhythm response state includes the systolic and diastolic blood pressure fluctuation amplitude offset, the heart rate fluctuation stability indicator, the blood oxygen and heart rate variability linkage characteristics, and the sampling frequency control mark. The heart-brain synchronous fluctuation combination group includes the consistency of the carotid artery pulsation delay direction, the blood oxygen gradient amplitude coordination, and the heart rate instantaneous change time series overlap. The joint change highlight segment includes the heart rate variability change trend, the blood oxygen saturation change trend, the joint change trend start and end time, and the combination association label. The response order arrangement includes the heart rate difference numerical sequence, the difference enhancement change trend, the body temperature amplitude comparison index, and the consistency of the change direction of the three physiological parameters. The cardiovascular and cerebrovascular disease risk assessment results include the time series overlap density, the fluctuation concentrated distribution segment, and the priority observation target labeling information.

[0019] See also Figure 2 , the rhythm sampling and tracking module includes: The pressure difference comparison submodule obtains the systolic and diastolic pressure values ​​of consecutive cycles, extracts the maximum and minimum differences between the two segments, subtracts them, and records the offset direction and change amplitude to generate a pressure difference offset sequence; First, the systolic and diastolic pressure data recorded in several consecutive heartbeat cycles are obtained. The difference between the systolic and diastolic pressures in each heartbeat cycle is extracted cycle by cycle to form a pressure difference value sequence. Then, two time windows are divided in the preset sampling segment. Usually, each window contains multiple consecutive heartbeat cycles. The maximum and minimum pressure differences are extracted in each time window in turn, and the fluctuation amplitude of the pressure difference in the segment is calculated respectively. Then, the difference of the pressure difference fluctuation amplitudes in the two time windows is calculated. This difference is the pressure difference offset amplitude, and the offset direction is marked according to the positive and negative difference. If the pressure difference amplitude of the second segment is greater than that of the first segment, it is defined as a positive offset, otherwise it is a negative offset. During execution, the minimum analysis period within each time window must not be less than five complete heartbeats to ensure the stability of the pressure difference extraction process. In addition, to prevent short-term abnormal data from affecting the offset judgment, each extracted pressure difference value is interval-limited. When a pressure difference value exceeds the physiological setting range (for example, the pressure difference is less than 20 mmHg or greater than 80 mmHg), the data is discarded. The judgment criteria for the pressure difference offset amplitude are: if the difference between the two segments is less than 3 mmHg, the offset is not recorded; when the difference is between 3 and 10 mmHg, it is marked as a mild offset; if it is greater than 10 mmHg, it is marked as a significant offset. This classification setting is based on the statistical acquisition of the pressure difference fluctuation range of the normal adult population and is set in combination with the actual characteristics of the blood pressure regulation rhythm. During the pressure difference offset sequence generation process, each difference calculation is completed, and the value and offset direction are stored in the sequence. The results of the most recent calculations are continuously recorded to form a time-continuous offset change sequence, which serves as the input basis for the subsequent heart rate screening submodule.

[0020] The heart rate fluctuation screening submodule calls the pressure difference offset sequence to detect whether the heart rate fluctuation in the sampling period is in a stable range. If it is satisfied, it collects blood oxygen and heart rate variability to obtain stable feature data; After receiving the pressure difference offset sequence, the first step is to determine whether there is a stable sampling basis. The basis for this determination is to evaluate the fluctuation amplitude of the heart rate interval sequence, and to use the calculated standard deviation or variation index of the interval data to determine whether the current heart rate is in a stable state. The specific steps are: intercept the heart rate interval data in a fixed-length time window, such as a continuous 30-second ECG record, and then extract the time value of each heartbeat interval (i.e., RR interval) from it. After forming a sequence, calculate its statistical standard deviation. If the standard deviation is less than the set stability threshold, the current heart rate fluctuation is considered acceptable. The stability threshold is usually set between 10 milliseconds and 15 milliseconds, depending on the age, body position, and activity status of the subject. Generally, for adults at rest, a standard deviation of less than 10 milliseconds can be judged as a stable heart rate. When this condition is met, the blood oxygen and heart rate variability (HRV) features are activated. Data collection process: Blood oxygen collection is carried out using the time synchronization method, with a synchronization time window of 10 seconds. 10 sampling points are collected and their average is taken as the blood oxygen level of the current segment. If any sampling value during the collection process is lower than 90%, the data for that segment is considered invalid; in the heart rate variability characteristics, the system extracts two key parameters including SDNN and RMSSD. SDNN reflects the overall interval fluctuation degree, and its value should be between 5 and 50 milliseconds. RMSSD reflects the rhythm changes of adjacent heartbeats, which is usually required to be between 10 and 40 milliseconds. If both parameters fall within the set range, the stable feature data is judged to be valid. During the entire screening process, when the heart rate fluctuation standard deviation exceeds the set upper limit, or the HRV feature parameter deviates from the preset range, the sampling data for that segment will not be adopted, and the system will re-execute the fluctuation screening in the next time period to ensure that the subsequent judgment link is based on high stability.

[0021] The response state determination submodule increases the sampling frequency based on the stable characteristic data and tracks whether the three-stage pressure difference offset changes return to the original range. The formula is: ; The combined fluctuation trend of the three-segment pressure difference offset value and the change rate is obtained by calculation to obtain the rhythm response state; in, is the rhythmic response state fluctuation, For the Segment pressure difference offset value, is the mean of the three-stage pressure difference offset, Indicates the The pressure difference deviation change rate of the segment is defined as the pressure difference between the front and rear segments divided by the corresponding sampling time difference. is the original pressure difference fluctuation mean; The response state judgment submodule increases the sampling frequency according to the stable characteristic data. The system increases the sampling frequency from the original Increase to , used to capture the pressure difference change process with higher accuracy. During the execution process after the sampling frequency is increased, the system extracts three sections of pressure difference offset values ​​according to the fixed sampling window length, and sets the sampling time of each section to , and extract the maximum difference between systolic and diastolic pressures in multiple heartbeat cycles in each segment to form the pressure difference offset value , the three sections are: ; First calculate the average value of the three pressure difference offset values: ; Then calculate the absolute value of the deviation between the pressure difference offset value and the mean value of each section (square root of the deviation): ; ; ; Calculate the absolute value of the pressure difference between each segment and the previous segment as the inter-segment jump amplitude: ; ; ; Add the deviation term of each segment to the inter-segment jump term: ; ; ; The sum of the three sub-items is: ; Calculate its average: ; The original reference pressure difference fluctuation mean is set as: ; Finally, the rhythm response state fluctuation is calculated: ; Conclusion Explanation: Final calculation results Greater than the upper limit of the stability range , so it is determined that the sampling segment is in a rhythmic unstable state.

[0022] See also Figure 3 , the wave pattern recognition module includes: The indicator extraction submodule obtains the carotid artery pulsation delay, blood oxygen gradient amplitude, and heart rate instantaneous change values ​​at the corresponding time point based on the marked sampling stage in the rhythm response state. It uniformly numbers the three indicators according to the sampling time sequence, establishes a cross-indicator time axis mapping relationship, and obtains the original sequence of the three fluctuations; First, the timestamp range of each sampling frame in this stage is parsed, and the matching carotid pulsation delay data, blood oxygen gradient amplitude data and heart rate instantaneous change data are extracted in sequence. Under the condition of a sampling frequency of 5Hz, 5 frames of data can be collected per second, so the system establishes a corresponding index with each frame as 1 numbering unit. In the actual operation example, the 10th to 12th seconds are marked as the abnormal response stage, and the corresponding frame numbers are 51 to 60 frames. The system extracts all three indicator values ​​within this time range. The carotid pulsation delay data is obtained by the time interval from the starting point of the ECG R wave to the maximum slope point of the rising segment of the carotid pulse waveform. The value range is usually between 110 and 150 milliseconds. If the continuous sampling values ​​are 122, 124, 127, 130, 131, and 129 milliseconds, the system is marked as frames 1 to 6 in frame order. The blood oxygen gradient amplitude data is obtained by the current time point Sp The value is obtained by subtracting the average of the previous five frames from the current value. If the previous average was 96.5% and the current value was 97.8%, the amplitude is 1.3%. Each amplitude calculation value must exceed 0.5% to be considered valid. This threshold is set based on the 95% confidence lower limit of the standard deviation of the basal blood oxygen stability interval of 50 healthy volunteers to ensure that measurement perturbations can be excluded. The instantaneous heart rate change value is converted from the RR interval change to the instantaneous heart rate and the difference is judged with the previous frame. If the previous frame is 78bpm and the current frame is 80bpm, the instantaneous change is +2bpm. After extracting the three data, the system uniformly numbers them. Even if there is a very small sampling time offset between the acquisition devices, the frame numbers are unified through timestamp matching, ultimately forming an aligned indicator original fluctuation sequence. Each indicator has data records in each frame and forms a three-column number mapping table for subsequent module calls.

[0023] The direction consistency judgment submodule calls the three original fluctuation sequences and compares the fluctuation directions of the three indicators within the same sampling period to determine whether they are all rising, falling, or stable. Data points with inconsistent directions are eliminated, and only time periods with consistent directions are retained to obtain the same-direction fluctuation intervals. After reading the three aligned and numbered original sequences of fluctuations, the change trends of the three indicators corresponding to each frame number are judged one by one. During the execution process, the system identifies the direction by the increase or decrease of the values ​​between adjacent frames. When the value of the next frame is higher than the current frame and exceeds the set threshold, it is marked as rising, and lower than the current frame, it is marked as falling. If the change amplitude is lower than the set direction judgment threshold, it is marked as stable. This threshold is set for each indicator: the carotid artery pulsation delay direction judgment threshold is 1 millisecond, the blood oxygen gradient amplitude direction judgment threshold is 0.1%, and the heart rate instantaneous change direction judgment threshold is 1bpm. This setting is based on the statistical variation range of 200 groups of actual clinical monitoring data to ensure that small fluctuations are not mistaken for trend changes. The system judges the change trend between frame 1 and frame 2 frame by frame, and the values ​​of the three indicators corresponding to frames 1 to 6 are 122 / The blood oxygen gradient amplitudes are 1.2 / 1.4 / 1.6 / 1.5 / 1.7 / 1.6%, with a trend of rising, rising, falling, rising, and falling. The heart rate instantaneous changes are 78 / 79 / 80 / 80 / 81 / 80 bpm, with a trend of rising, rising, stable, rising, and falling. The system tags each frame with three trend labels and checks whether the three directions are consistent within the frame according to the consistency standard. Only frame numbers with consistent three trends are retained. In this example, frame 2 has all three indicators rising, so it is retained. Frame 4 has two rising indicators but the blood oxygen level is falling, so it is discarded. The system finally records frames with consistent directions as frames 23, 56, etc., and internally generates time period identifiers and frame range indexes to exclude segments with inconsistent trend directions.

[0024] The synchronous combination screening submodule identifies whether the sampling points of indicators overlap on the time axis based on the marked time periods in the same-direction fluctuation interval, and screens indicator combinations with consistent directions and synchronized sampling points to generate a heart-brain synchronous fluctuation combination group; Call the index of the time period with consistent direction obtained by the above screening, check whether the three indicators in each frame have valid sampling at the same time, and compare the accuracy of the sampling time to determine whether the timestamps of the three indicators in each frame are synchronized within the set error tolerance. The error tolerance is set to 0.1 seconds, that is, if the maximum difference between the sampling times of the three indicators does not exceed 100 milliseconds, they are considered synchronized. During the synchronization check, the system extracts the sampling timestamp of each frame from the frame number table. For example, the sampling times of the three indicators in frame 2 are 10.20s, 10.18s, and 10.25s respectively, and the maximum difference is 0.07s, which is less than 0.1 seconds, which meets the synchronization conditions. The sampling times of frame 5 are 12.04s, 12.10s, and 12.09s, with a maximum difference of 0.06 seconds, also meets the conditions. If any indicator of a frame is missing or the time difference exceeds the threshold, the frame is marked as out of sync and eliminated. The synchronized frame number is intersected with the previous consistency judgment to filter out the final valid frame set. The system combines these numbered frames into a fluctuation combination group. The group must have at least 3 consecutive frames, and each frame must meet the two conditions of consistent direction and synchronous sampling. The frame numbers that meet the conditions will be recorded as valid groups by the system. For example, the three frames in frames 2 to 4 have consistent directions and synchronous sampling, forming the first fluctuation combination group and named "Group-A". The group number, start and end frame numbers, and the three indicator values ​​are recorded to construct an output table for subsequent structural evaluation.

[0025] See also Figure 4 , the abnormal joint variation annotation module includes: The variation index extraction submodule extracts the heart rate variability and blood oxygen saturation data within the corresponding sampling stage based on the combinations in the heart-brain synchronization fluctuation combination group, uniformly numbers the time periods, establishes a mapping relationship between the indicators and the sampling segments, and generates a variation index time series; Based on the heart-brain synchronization fluctuation combination group, the two indicators of heart rate variability and blood oxygen saturation in the corresponding time period are extracted from each confirmed synchronization group. The extraction operation is based on the time axis number to ensure that the sampling sequence is continuous and traceable. In the actual implementation process, assuming that Group-A covers frame numbers 1 to 6, the system obtains the original values ​​of the two indicators under each frame number in turn. Heart rate variability is a statistical ECG value once per second, reflecting the degree of RR interval fluctuation in milliseconds. Blood oxygen saturation takes the current Sp The instantaneous reading is expressed as a percentage. The sampling frequency matches the heart rate variability. The system uses the frame number as the primary key to construct a mapping relationship between indicators, such as frame number 1: heart rate variability 52.3ms, blood oxygen saturation 97.1%; frame number 2: 53.8ms, 97.3%; frame number 3: 55.6ms, 97.5%; frame number 4: 55.4ms, 97.4%; frame number 5: 54.0ms, 97.2%; and frame number 6: 53.0ms, 97.0%. This mapping relationship is recorded in the sequence database as a structured table. To ensure that the sampled data is within the normal range, the system has preset rejection rules: if the heart rate variability value is less than 25ms or greater than 120ms, it is considered an extreme value and is directly rejected. If the blood oxygen saturation is less than 94% or greater than 100%, it is considered unstable and the data is also rejected. The threshold setting is based on the reference range for the human body at rest specified in the national standard GB9706, and abnormal intervals are supplemented by clinical basic monitoring data of 100 people. The above two parameters need to be recorded uniformly when the time number is complete and the data value is reasonable, and finally form a time series of heart rate variability and blood oxygen saturation for subsequent call.

[0026] The trend consistency judgment submodule calls the variation index time series to judge the change direction of heart rate variability and blood oxygen saturation values ​​in continuous sampling segments, selects continuous stages with consistent fluctuation directions, establishes start and end time indexes, and obtains trend consistency interval information; Calling the generated time series of variability indicators, starting with number 1, analyzes the consistency of change in heart rate variability and blood oxygen saturation within consecutive numbered segments. The judgment criteria are: if heart rate variability and blood oxygen saturation increase between the current and next numbered segments, it is marked as rising consistency; if both decrease, it is marked as falling consistency; if the change amplitude does not exceed the threshold, it is marked as stable consistency; if the direction is opposite, the segment is discarded. In specific implementation, the system sets the judgment threshold as follows: heart rate variability changes less than 0.5ms are considered stable, and blood oxygen saturation changes less than 0.1% are also considered stable. This threshold is derived from the results of 120 continuous samplings of static subjects over 20 minutes. The standard deviation of the maximum fluctuation amplitude is 0.43ms and 0.08%, so the slightly higher value is used as the actual standard. In the sample, from 1 to 2, heart rate variability increased from 52.3ms to 53.8ms, and blood oxygen increased from 97.1% to 97.3%, both in an upward direction; from 2 to 3, heart rate variability increased from 53.8ms to 55.6ms, and blood oxygen increased from 97.3% to 97.5%, both showing an upward trend; from 3 to 4, heart rate variability decreased slightly to 55.4ms (-0.2ms), and blood oxygen decreased to 97.4% (-0.1%), both within the stable threshold range, and therefore also recorded as consistent; from 4 to 5, heart rate variability decreased to 54.0ms (-1.4ms), and blood oxygen decreased to 97.2% (-0.2%), both showing a consistent trend; from 5 to 6, heart rate variability decreased to 53.0ms (-1.0ms), and blood oxygen decreased to 97.0% (-0.2%), both showing a downward trend. The entire segment from 1 to 6 is a complete trend-consistent segment. The system uses this segment-by-segment direction judgment method to screen out all number sequences with consistent continuous directions, record their starting and ending numbers, and build a list of trend-consistent interval information for use in the next module.

[0027] The highlight segment construction submodule integrates the combination content, start and end time periods, and grouping based on the trend consistency interval information, classifies and labels the stage behaviors that meet the conditions, and records them in the database to obtain the joint change highlight segments; The system receives trend-consistent interval information and combined group information, and cross-checks each trend-consistent segment to see if it completely falls within any synchronization group. If a complete inclusion relationship is satisfied, the segment is extracted as a candidate highlight segment. The system uses the start and end numbers of the trend-consistent segment as the basis, combined with the group ID to which it belongs, to perform a combination labeling and classify the segment behavior. During execution, the system assigns a unique ID to each candidate highlight segment, such as number P01, corresponding to group Group-A, and number segments 1 to 6. The content written into the data structure includes fields such as the segment number, group number, number segment start and end, heart rate variability average, blood oxygen saturation average, direction label, and number of data points. For example, in number segments 1 to 6, the heart rate variability average is 54.0ms, the blood oxygen average is 97.4%, the direction label is consistent from rising to falling, and the number of data points is 6, which meets the system's preset minimum highlight segment length of not less than 3 frames. The lower threshold of 3 frames is based on the variation in segment identification accuracy observed during system testing. When the segment length is less than 3 frames, the fluctuation trend is not significantly different from the background fluctuation. Therefore, a minimum of three frames is required for the segment to be recorded as valid. The system ultimately writes the constructed highlighted segment information into a database structure table and associates it with the corresponding group, enabling independent storage of the highlighted segments and interfacing with the query interface, forming the final set of linked highlighted segments.

[0028] See also Figure 5 , the priority follow-up sorting module includes: The difference sequence construction submodule extracts the maximum and minimum heart rate values ​​based on the joint variation highlight segment, calculates the difference between the two, constructs the difference record corresponding to each segment, and then arranges them in sampling time sequence to form a complete sequence to obtain the heart rate difference sequence; After receiving the linked-change highlighted segment data, the system iterates through each segment, analyzing all heart rate data within that sampling timeframe. Within each segment, the system extracts the maximum and minimum heart rate values. The maximum is the highest value among all the heart rate samples in that segment, and the minimum is the lowest value within that segment. The difference between the two is the heart rate difference for that segment. For example, in segment 1, the consecutive heart rate samples are 84, 86, 89, 92, 95, and 98, with a maximum of 98 and a minimum of 84, resulting in a difference of 14. In segment 2, the heart rate values ​​are 88, 90, 93, 97, and 101, with a difference of 13. In segment 3, the heart rate values ​​are 85, 87, 89, 93, and 97, with a difference of 12. In segment 4, the heart rate values ​​are 91, 94, 97, 100, and 103, with a difference of 12. In segment 5, the heart rate values ​​are 89, 91, 94, 97, and 100, with a difference of 11. The system records the differences calculated above in the order of segment numbers, constructing a difference sequence of 14, 13, 12, 12, and 11, and synchronously attaches this sequence to the corresponding segment data to establish a complete segment number and heart rate difference mapping structure. During the construction process, the system sets two elimination rules: first, when the number of heart rate data in a certain segment is less than three points, no record is made; second, when the difference between the maximum and minimum values ​​is less than 5bpm, the fluctuation is considered unrepresentative and is eliminated. This value is set based on the statistical range of heart rate differences among people at rest. In actual monitoring, it was found that the heart rate variation range of more than 95% of people at rest is not less than 6bpm, so 5bpm is set as the minimum acceptance threshold. After the screening is completed, the difference sequence data will serve as the input basis for the next stage of body temperature comparison analysis.

[0029] The temperature fluctuation comparison submodule calls the heart rate difference sequence, synchronously extracts the segment temperature variation amplitude, and compares whether the temperature variation amplitude shows a synchronous strengthening trend with the current heart rate difference amplitude. The formula is:

[0030] The calculation obtains the joint deviation trend of each segment of heart rate difference and body temperature fluctuation, selects the sampling segments with consistent trends, and generates the trend synchronization segment index; in, Indicates the trend synchronization deviation, 、 Respectively The maximum and minimum heart rate values ​​of the segment, Indicates the The temperature variation range of the segment, Indicates the Average body temperature of the segment, is the total number of sample segments, Represents the average heart rate; The temperature fluctuation comparison submodule calls the heart rate difference sequence and extracts the segment body temperature variation value synchronously. First, in each sampling segment Extract the maximum heart rate of this segment With minimum value , based on which the heart rate difference is calculated , and then calculate the average heart rate as , based on which the relative heart rate difference is obtained , and then obtain the temperature variation amplitude of this section Average body temperature , multiply the relative difference in heart rate by the body temperature amplitude, and then divide it by the mean body temperature to get the trend value of this section , for all sampling segments Take the average to get the trend synchronization deviation: ; The following calculation is performed using three sampling segments as an example: Paragraph 1: , ,but , , , body temperature fluctuation , mean body temperature ,but ; Paragraph 2: , ,but , , , , , ; Paragraph 3: , ,but , , , , , ; The average of the three results is the trend synchronization deviation: ; Set the trend consistency judgment threshold to ,but , indicating that the current three sampling segments are trend synchronized and can all be filtered into the trend synchronization segment index.

[0031] The multi-parameter direction analysis submodule synchronizes the segment index with the trend, analyzes the direction of change of the heart rate, blood pressure, and blood oxygen in the segment, and determines whether the three are rising, falling, or stable at the same time. If they are consistent, the combined features are recorded and the response ranking level of the current segment is marked to obtain the response order arrangement; The sampling sequence of the three indicators of heart rate, blood pressure, and blood oxygen is read segment by segment according to the trend synchronization segment index, and the change trends of the three indicators are judged to be consistent. The system sets the judgment criteria as follows: if the indicator has more than two consecutive increases within the sampling segment and the increase is greater than the basic set threshold, it is recorded as an increase; if it has continuously decreased and the decrease meets the set threshold, it is recorded as a decrease; if the change value is less than the threshold, it is recorded as stable. The thresholds of the three indicators are set at 2bpm for heart rate, 3mmHg for blood pressure, and 0.2% for blood oxygen. This setting is based on the sampling accuracy of the medical equipment and the standard deviation measurement results of 300 test samples. Taking segment 2 as an example, the heart rate sampling sequence is 88, 90, 93, 97, and 101, with a continuous increase of more than 2 bpm, which is considered rising. The blood pressure sequence is 113, 115, 118, 122, and 126, with each increase greater than 3 mmHg, which is considered rising. The blood oxygen data is 96.5, 96.7, 96.9, 97.0, and 97.2, with each increase of more than 0.2%, which is considered rising. These three consistent values ​​are marked as "consistent rising," and this segment is identified as a consistent direction response segment. Looking at segment 4, the heart rate is 91, 94, 97, 100, and 103, with an upward trend. The blood pressure is 125, 123, 121, 119, and 117, with a continuous decrease. The blood oxygen is 96.8, 96.8, 96.7, 96.6, and 96.6, with a change of less than 0.2%, which is considered stable. The three values ​​in this segment are inconsistent in direction and are therefore excluded. In segment 5, the heart rate fluctuated between 11 and 13, the blood pressure rose slightly between 110 and 114, and the blood oxygen level remained at 96.9, but fluctuated within 0.2%. The heart rate was rising, the blood pressure was stable, and the blood oxygen level was stable, failing to meet the three criteria for consistency. This segment was deemed inconsistent and removed. The system sorted the remaining segments based on directional consistency. Consistent responses from all three criteria were marked as Level I, consistent responses from any two criteria were marked as Level II, and consistent responses from only one or no criteria were marked as Level III. Finally, segment 2 was designated a Level I response and entered into the response level output table.

[0032] See also Figure 6 , the risk location classification module includes: The first segment reading submodule extracts the time information corresponding to the first response segment based on the response order, verifies the start and end time and index position, and constructs a mapping relationship between the segment and the entire time series to obtain the time range of the first response; First, determine the top response segment according to the response order. After obtaining the response segment number, call the time tag data corresponding to the segment immediately. Get the actual corresponding time range by calling the start time and end time fields recorded in the original sampling record. Then call the index position corresponding to the segment in the global time series and convert it by the sampling frequency to obtain its start and end sampling point numbers. For example, if the sampling frequency is once per second, the time covered from 08:00 to 08:20 is 1200 seconds, corresponding to 1200 sampling points. If the starting sampling point is 1000, the ending sampling point is 2200. Then, By constructing a mapping record of the start and end index positions of the response segment in the entire time series, a mapping table of the response segment and the time series is formed. The mapping table records fields such as the response segment number, start and end time, start and end index, and duration, which are further used for subsequent cross-matching operations. For example, if the first response segment is numbered S1, it ranks first in the response sequence table, and the corresponding time label is 08:00 to 08:20, then it is located at the index position 1000 to 2200 in the original sampling data. This information can be directly obtained by reading the response list header and its corresponding time identifier, thereby forming the first response time range data structure.

[0033] The overlap segment judgment submodule calls the first response time range and cross-matches it with the time periods of all combinations in the heart-brain synchronization fluctuation combination group to screen whether there are multiple combinations that repeatedly appear within the time range covered by the response segment. If the combinations highly overlap in the target time period, they are identified as overlap concentration intervals. First, retrieve the first response time range information obtained in the previous stage. This information includes the start time, end time, and sampling index position of the response segment. Then, retrieve the start and end time information of all combinations in the heart-brain synchronization fluctuation combination group. By traversing the start time and end time of all combination segments one by one, cross-judgment is performed with the time range of the current response segment. The judgment rule is that the end time of the combination segment must be greater than or equal to the start time of the response segment, and the start time of the combination segment must be less than or equal to the end time of the response segment. If the judgment condition is met, there is a time overlap between the combination segment and the response segment. Subsequently, the overlap ratio of all overlapping combination segments is further calculated, that is, the length of the overlap time of the combination segment within the response segment range is divided by the original duration of the combination segment to set the overlap ratio. The threshold is 0.6, which is derived from the sensitivity analysis results of overlapping segments to disease evolution in a large number of historical patient data samples. It is confirmed that when the overlap ratio exceeds 60%, the clustering characteristics of the combination segment are significantly enhanced. For example, the duration of combination G1 is 15 minutes. If it overlaps with the response segment for 13 minutes between 08:00 and 08:20, its overlap ratio is 13 / 15=0.867, which is significantly higher than the set threshold. Therefore, it is identified as a highly overlapping combination segment. The calculation process is repeated to screen all combinations in the combination group, and the number of all combination segments that meet the high overlap and their start and end time distribution are counted. If the number of overlapping segments is not less than 3 and the total coverage time is not less than 10 minutes, the time of the current response segment is marked as the overlapping concentrated interval.

[0034] The centralized segment labeling submodule performs priority labeling based on the time series information and combination quantity distribution of the overlapping centralized intervals, and simultaneously records the combination type, coverage and distribution characteristics to obtain cardiovascular and cerebrovascular disease risk assessment results; After identifying the overlapping concentrated interval, the combination within the time period is labeled. First, the list of combination segment numbers that meet the overlap conditions and their start and end times, combination types, durations, and overlap degree data are retrieved. Then, the combination types are classified and processed. The combinations are divided into three categories according to their dominant parameters: heart rate-dominated, blood oxygen-dominated, or cross-fluctuation. The number of each type of combination is counted separately. If a certain type of combination accounts for more than 50%, it is used as the dominant type of the current segment. Secondly, the start and end times of each combination segment are sorted to determine its distribution density within the segment, that is, the number of combination segments per unit time. If If the number of combinations exceeds 4 within 10 minutes, it is marked as a high-density segment. The segment is then prioritized based on the combination type distribution, average overlap, and density level. For example, if combinations G1 to G4 are all cross-fluctuation types, appear between 08:00 and 08:20, and the overlap exceeds 85%, then the segment is labeled as a "high-density cross-fluctuation overlap segment" with a high-level annotation level. The final summary record fields include: segment start and end time, combination type distribution, dominant type, number of overlap segments, average overlap ratio, distribution density, and annotation level, thereby generating the final cardiovascular and cerebrovascular risk assessment output information.

[0035] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. Cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data, characterized by: The system comprises: The rhythm sampling and tracking module collects continuous systolic and diastolic blood pressure, calculates the difference between the two segments, determines the direction and amplitude of changes, and synchronously calls blood oxygen and heart rate variability when the heart rate is stable and increases the sampling frequency. After tracking three segments, it determines whether there is regression and generates a rhythm response state. The fluctuation pattern recognition module calls the carotid artery delay, blood oxygen gradient, and heart rate variation according to the rhythm response state, determines whether the change directions of the three are consistent, selects combinations with consistent directions and overlapping time sequences, and generates a heart-brain synchronization fluctuation combination group; The abnormal joint variation annotation module extracts the heart rate variability and blood oxygen data of the heart-brain synchronous fluctuation combination group, determines whether they show a consistent trend, and if so, marks the time and combination, classifies and records them, and generates a joint variation highlight segment; The priority follow-up sorting module extracts the heart rate difference sequence of the joint change highlight segment, determines whether the trend is strengthening, combines the body temperature change amplitude, analyzes the direction of indicator change, and generates a response order arrangement; The risk location division module reads the first response segment in the response order arrangement, cross-compares the combined group segments, and determines whether they are concentratedly distributed. If so, it is marked as a priority observation target to generate a cardiovascular and cerebrovascular disease risk assessment result.

2. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 1 is characterized in that: The rhythm response state includes the systolic and diastolic pressure fluctuation amplitude offset, the heart rate fluctuation stability mark, the blood oxygen and heart rate variability linkage characteristics, and the sampling frequency control mark. The heart-brain synchronous fluctuation combination group includes the consistency of the carotid artery pulsation delay direction, the blood oxygen gradient amplitude coordination, and the heart rate instantaneous change time series overlap. The joint change highlight segment includes the heart rate variability change trend, the blood oxygen saturation change trend, the joint change trend start and end time, and the combination association label. The response order arrangement includes the heart rate difference numerical sequence, the difference enhancement change trend, the body temperature amplitude comparison index, and the consistency of the change direction of the three physiological parameters. The cardiovascular and cerebrovascular disease risk assessment results include the time series overlap density, the fluctuation concentrated distribution segment, and the priority observation target labeling information.

3. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 1 is characterized in that: The combination of consistent direction and overlapping time sequence refers to a combination of physiological indicators with the same changing trend and synchronized changing time points within the same time period; The indicator change direction refers to the trend direction of the physiological parameter showing continuous increase, decrease and stability over a period of time.

4. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 1 is characterized in that: The rhythm sampling and tracking module includes: The pressure difference comparison submodule obtains the systolic and diastolic pressure values ​​of consecutive cycles, extracts the maximum and minimum differences between the two segments, subtracts them, and records the offset direction and change amplitude to generate a pressure difference offset sequence; The heart rate fluctuation screening submodule calls the pressure difference offset sequence to detect whether the heart rate fluctuation in the sampling period is in a stable range. If so, the blood oxygen and heart rate variability are collected to obtain stable feature data; The response state determination submodule increases the sampling frequency according to the stable characteristic data, tracks whether the three-stage pressure difference offset changes return to the original range, calculates and obtains the joint fluctuation trend of the three-stage pressure difference offset value and the change rate, and obtains the rhythm response state.

5. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 4 is characterized in that: The wave pattern association module includes: The indicator extraction submodule obtains the carotid artery pulsation delay, blood oxygen gradient amplitude, and heart rate instantaneous change values ​​at the corresponding time point based on the marked sampling stage in the rhythm response state, uniformly numbers the three indicators according to the sampling time sequence, establishes a cross-indicator time axis mapping relationship, and obtains the original sequence of the three fluctuations; The direction consistency judgment submodule calls the three original fluctuation sequences, compares the fluctuation directions of the three indicators within the same sampling period, and determines whether they are all rising, falling, or stable. Data points with inconsistent directions are eliminated, and only time periods with consistent directions are retained to obtain the same-direction fluctuation interval; The synchronous combination screening submodule identifies whether there is overlap in the sampling points of the indicators on the time axis according to the marked time periods in the same-direction fluctuation interval, screens indicator combinations with consistent directions and synchronized sampling points, and generates a heart-brain synchronous fluctuation combination group.

6. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 5 is characterized in that: The abnormal joint change marking module includes: The variation index extraction submodule extracts the heart rate variability and blood oxygen saturation data in the corresponding sampling stage based on the combinations in the heart-brain synchronization fluctuation combination group, uniformly numbers the time periods, establishes a mapping relationship between the indicators and the sampling segments, and generates a variation index time series; The trend consistency judgment submodule calls the variation index time series, judges the change direction of heart rate variability and blood oxygen saturation value in the continuous sampling segment, selects the continuous stages with the same fluctuation direction, establishes the start and end time indexes, and obtains the trend consistency interval information; The highlighted segment construction submodule integrates the combination content, start and end time periods, and grouping according to the trend consistent interval information, classifies and labels the stage behaviors that meet the conditions, and records them into the database to obtain the joint change highlighted segments.

7. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 6 is characterized in that: The priority follow-up sorting module includes: The difference sequence construction submodule extracts the maximum and minimum heart rate values ​​based on the joint change highlighted segments, calculates the difference between the two, constructs the difference records corresponding to each segment, and then arranges them in sampling time sequence to form a complete sequence to obtain the heart rate difference sequence; The temperature fluctuation comparison submodule calls the heart rate difference sequence, synchronously extracts the segmented body temperature variation amplitude, compares whether the body temperature variation amplitude shows a synchronous enhancement trend with the current heart rate difference amplitude, calculates the joint offset trend degree of each segment of heart rate difference and body temperature fluctuation, selects sampling segments with consistent trends, and generates a trend synchronization segment index; The multi-parameter direction analysis submodule synchronizes the segment index according to the trend, analyzes the change direction of the heart rate, blood pressure, and blood oxygen in the segment, and determines whether the three are rising, falling, and stable at the same time. If they are consistent, the combined features are recorded and the response sorting level of the current segment is marked to obtain the response order arrangement.

8. The cardiovascular and cerebrovascular disease risk assessment and management system based on accurate health data drive according to claim 7 is characterized in that: The risk location classification module includes: The first segment reading submodule extracts the time information corresponding to the first response segment based on the response order arrangement, verifies the start and end time and index position, and constructs a mapping relationship between the segment and the entire time series to obtain the time range of the first response; The overlap segment judgment submodule calls the first response time range and cross-matches it with the time periods of all combinations in the heart-brain synchronization fluctuation combination group to screen whether there are multiple combinations that repeatedly appear within the time range covered by the response segment. If the combinations highly overlap in the target time period, they are identified as overlap concentration intervals. The centralized segment labeling submodule performs priority labeling according to the time series information and combination quantity distribution of the overlapping centralized intervals, and simultaneously records the combination type, coverage and distribution characteristics to obtain the cardiovascular and cerebrovascular disease risk assessment results.

Images