A health management and physical examination big data platform

Abstract

This invention relates to the field of health management technology, specifically to a health management physical examination big data platform. The system includes a fluctuation start point localization module, a rhythm structure adjustment module, a trend trigger identification module, a vital sign linkage extraction module, and an intervention task sequencing module. In this invention, the fluctuation start point is located by jointly locating the direction of resistance change and abrupt shifts in offset. The sampling interval is dynamically adjusted and the fluctuation sequence is rearranged. The trend continuation path is identified using boundary deviation states, enhancing the coherence and stability of trend identification. The response start point and continuation region are extracted by combining directional consistency items in respiratory vital signs, ensuring the completeness of time-domain response coverage. Furthermore, a priority calling order is constructed based on the distribution density, frequency of occurrence, and duration of vital signs on the time axis. This achieves accurate identification of trend start points, orderly adjustment of rhythm structures, and optimization of intervention task timing in the dynamic linkage process of multiple vital signs, improving response efficiency and control accuracy in the evolution of health status.

Description

Technical Field

[0001] This invention relates to the field of health management technology, and in particular to a health management physical examination big data platform. Background Technology

[0002] The field of health management technology involves a comprehensive service system that uses information technology to dynamically monitor, assess, and intervene in the health status of individuals or groups. Its core aspects include health data collection, vital sign analysis, risk prediction and assessment, health intervention program development, and health behavior management. This field is based on medical informatics and integrates cutting-edge technologies such as big data, artificial intelligence, and wearable devices to achieve systematic management of population health throughout its entire life cycle. The development of this technology stems from the public health pressures brought about by strained medical resources and the high incidence of chronic diseases, requiring the establishment of a data-driven proactive health management mechanism to improve disease prevention efficiency and personalized service levels. Traditional health management physical examination data platforms refer to information systems that record and classify health status based on collected individual physical examination data. These platforms primarily address the technical issues of complex and inconsistent physical examination data sources and fragmented health management decision-making bases. Traditional patent themes typically address this by storing physical examination data in tabular form and manually classifying and summarizing it. The data structure relies on database field rules, and analysis usually involves manual classification and comparison based on preset query conditions. Individual health status assessments are based on static standard models or are completed by health management personnel based on experience.

[0003] Traditional physical examination data platforms suffer from a lag in identifying the initial state when dealing with rapidly evolving dynamic signal sequences. Preset field frameworks limit the flexible expression of continuously changing parameters, and manual classification methods struggle to reflect the continuity and directionality of vital signs trends over time. When multiple fluctuation phases exist in the sampling sequence, the platform struggles to determine their temporal sequence and trend continuity, leading to deviations in signal change path judgment. The lack of trend matching criteria in the linkage judgment of vital signs such as respiration also easily causes ambiguity in response segment identification. In terms of intervention triggering, there are problems with disordered response order and delayed call timing, reducing the matching degree and execution efficiency in the vital sign intervention process. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a health management and physical examination big data platform.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A health management and physical examination big data platform includes:

[0007] The fluctuation start point localization module acquires tissue resistance records for a time period in the arterial ohmogram detection scenario, determines the direction of adjacent resistance changes, compares the difference between the change position of the interface and the electrical wave offset, selects the deviation time point, and moves it backward as the new sampling start point to obtain the fluctuation delay start time point;

[0008] The rhythm structure adjustment module sets a starting point based on the start time of the fluctuation delay, compares the current and previous fluctuation directions, and if they are consistent, adjusts the segment length and rearranges the order of fluctuation points to obtain the rhythm adjustment interval.

[0009] The trend trigger recognition module, based on the rhythm adjustment interval, calibrates the blood pressure boundary position, determines whether the measuring point continuously falls outside the boundary, and makes a continuous judgment on the change path and interval to obtain the arterial pressure change trend recognition result.

[0010] The vital sign linkage extraction module, based on the trend direction in the arterial pressure change trend recognition result, selects items that are consistent with the changes in expiratory time, inspiratory holding time and lung volume, compares the initial response position and the coverage of the continuous segment, and obtains the associated vital sign response trajectory.

[0011] As a further aspect of the present invention, the fluctuation delay start point includes reference time positioning, offset distance quantification, and updated sampling start point; the rhythm adjustment interval includes time range adjustment, fluctuation sequence rearrangement, and change trend matching; the arterial pressure change trend identification result includes exceeding the measurement point boundary, continuous deviation segment, and trend path structure; and the associated vital sign response trajectory includes consistent change start point, response duration interval, and fluctuation period coverage.

[0012] As a further aspect of the present invention, the fluctuation start point positioning module includes:

[0013] The resistance trend difference detection submodule acquires continuous time period tissue resistance records in the arterial ohmogram detection scenario, extracts the resistance change trend of each time period in the same time series, compares the direction markers of resistance change in any two adjacent time periods, identifies the time period positions with inconsistent directions, and obtains the intervals with inconsistent resistance directions.

[0014] The junction offset anomaly detection submodule extracts the corresponding junction position state and radio wave offset state sequence based on the interval of inconsistent resistance direction, compares the junction state change amplitude with the set junction change recognition threshold, filters the time period position of the offset state anomaly, and generates a junction offset abrupt change position set.

[0015] The fluctuation start point positioning submodule, based on the position index of the time period of the junction offset change position set, locates the time point sequence in the resistor record, identifies the position of the offset state change, and adjusts the time point backward by a specified amount of time as the new sampling start point to obtain the fluctuation delay start time point.

[0016] As a further aspect of the present invention, the rhythm structure adjustment module includes:

[0017] The direction coherence recognition submodule extracts the sampling sequence within the current fluctuation segment based on the starting position of the fluctuation delay start point, records the change direction of the fluctuation range corresponding to the time point, extracts the change direction marker within the previous segment, compares the consistency between the current segment and the previous segment in the corresponding time direction, filters out continuous time periods with consistent direction, and obtains the direction coherence interval.

[0018] The segment boundary revision submodule identifies the degree of overlap between the start and end times of the current fluctuation segment and the coverage area based on the time boundary of the directional continuous interval, and determines whether it meets the conditions for extension and compression. If it does, the start and end times of the current fluctuation segment are revised to generate the adjusted segment range.

[0019] The rhythm reordering submodule calls the revised time range in the adjusted segment range, reorders the contained fluctuation points according to the original fluctuation direction, and arranges the sorted fluctuation point position sequence as the segment benchmark to obtain the rhythm adjustment interval.

[0020] As a further aspect of the present invention, the trend trigger recognition module includes:

[0021] The boundary extraction and calibration submodule extracts the upper and lower blood pressure boundaries of three adjacent detection segments based on the signal sequence within the rhythm adjustment interval, calibrates the time nodes of the boundary values ​​in each segment, records the index sequence of the boundary time points in the segments, and generates a boundary position index set.

[0022] The boundary crossing interval filtering submodule calls the corresponding positions of the upper and lower limits of the boundary in the boundary position index set, extracts the position values ​​of the measurement points in the same segment, compares the positional relationship between each measurement point and the corresponding boundary value, and filters the time period positions where consecutive measurement points exceed the upper and lower boundaries to obtain the continuous boundary crossing point intervals.

[0023] The trend consistency determination submodule extracts the change path direction markers between measuring points and the time interval sequence between adjacent measuring points based on the position information of the measuring points in the continuous boundary point interval, determines whether the change direction is consistent and the interval does not exceed the trend continuity threshold, and obtains the arterial pressure change trend recognition result.

[0024] As a further aspect of the present invention, the vital sign linkage extraction module includes:

[0025] The direction matching and filtering submodule extracts the change direction markers of the sign sequences in the respiratory-related signs based on the trend information in the arterial pressure change trend recognition results. It then matches and filters the change direction of each sign with the trend direction, retains the sign data sequences with consistent directions, and obtains the direction-matched signs.

[0026] The morphological response segment extraction submodule calls the direction matching vital signs, identifies the location index of the direction change, records the position number of the starting point of the vital sign change in the original sampling sequence, determines whether there is a change trajectory in the continuous time period after the number that exceeds the morphological recognition benchmark value, and obtains the morphological response segment.

[0027] The response trajectory generation submodule extracts the covered time interval based on the start and end time index of the morphological response segment, calculates the fluctuation frequency value within the response segment, marks the distribution structure of the fluctuation interval within the time interval, and obtains the associated vital sign response trajectory.

[0028] As a further aspect of the present invention, the formula for calculating the fluctuation frequency value within the response segment is as follows:

[0029] ;

[0030] in, Representing the Fluctuation frequency value within each response segment Representing the The number of time index points within each response segment Representing the Within the first response segment The difference between the vital sign value at each time point and the vital sign value at the previous time point, i.e. , Representing the The standard deviation of phenotypic values ​​within each response segment Representing the The average value of the symptom within each response segment Representing the The maximum value of the eigenvalue within each response segment Representing the The minimum value of the eigenvalue within each response segment Representing the Within the first response segment Vital signs at each time point.

[0031] As a further aspect of the present invention, an intervention task sorting module is also included:

[0032] The intervention task sorting module identifies changes, durations, and frequencies of occurrence of vital signs in the associated vital sign response trajectory, and combines the intervention order to obtain a list of vital sign intervention execution order.

[0033] The list of intervention sequence for vital signs includes the order of appearance of vital signs, duration of changes, and frequency of changes.

[0034] As a further aspect of the present invention, the intervention task sorting module includes:

[0035] The vital sign first-order sorting submodule extracts the first response position of the vital sign on the sampling time axis based on all the responded vital signs in the associated vital sign response trajectory, numbers and sorts the first and first response points of the differentiated vital signs in chronological order, records the time index arrangement structure of each vital sign, and generates the vital sign time series order value.

[0036] The vital sign duration segment calculation submodule calls the time index of the vital sign in the sequential value of the vital sign time series, extracts the start and end boundaries of the continuous change of the vital sign in the sampling sequence, calculates the duration value of the change, and marks the complete segment index during the change period to obtain the duration interval of the vital sign change.

[0037] The integrated sorting and list generation submodule calculates the frequency of each vital sign change in the entire sequence based on the index of the change segment of vital signs in the duration interval of the vital sign change. It then uses time order, duration, and frequency of occurrence as sorting criteria to establish a multi-factor combination judgment logic and obtain a list of vital sign intervention execution order.

[0038] As a further aspect of the present invention, the formula for calculating the duration of the change is specifically as follows:

[0039] ;

[0040] in, Representative number is physical signs The duration of change during a certain continuous change process. Indicating physical signs The Index of the end time point of each change segment. Indicating physical signs The Index of the starting time point of each change segment. Indicating physical signs In the Within the first change segment The difference in vital sign values ​​between each sampling point and its previous sampling point Indicating physical signs In the Within the first change segment The time difference between each sampling point and its previous sampling point Indicating physical signs In the The average rate of change of all sampling point pairs within each change segment. Indicating physical signs In the Number of sampling point pairs within each variation segment This is the index for the difference sequence number.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0042] In this invention, the starting point of fluctuation is located by combining the direction of resistance change and the abrupt shift in offset. The sampling interval is dynamically adjusted and the fluctuation sequence is rearranged. The trend continuation path is identified by using the boundary deviation state, which enhances the coherence and stability of trend identification. The response starting point and continuation region are extracted by combining the directional consistency items in respiratory signs to ensure the integrity of time-domain response coverage. Furthermore, a priority calling order is constructed based on the distribution density, frequency of occurrence and duration of signs on the time axis. This enables accurate identification of trend starting points, orderly adjustment of rhythm structure and optimization of intervention task timing in the dynamic linkage process of multiple signs, thereby improving the response efficiency and control accuracy in the evolution of health status. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a system flowchart of the present invention.

[0045] Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0046] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0047] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0048] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0049] Please see Figure 1 and Figure 2 A health management and physical examination big data platform includes:

[0050] The fluctuation start point localization module acquires tissue resistance records for three consecutive time periods in the arterial ohmogram detection scenario, determines whether the resistance change direction is consistent between any two adjacent time periods, compares the deviation between the interface change position and the electrical wave offset state execution position, filters out the reference time position corresponding to the significant change item, and moves this time position backward as the new sampling start point to obtain the fluctuation delay start time point.

[0051] The rhythm structure adjustment module, based on the starting position of the fluctuation delay start point, determines whether the trend of the sampling sequence in the fluctuation range within the current fluctuation segment is consistent with the direction of the previous segment. If the direction of change is maintained, the current time range is extended or compressed accordingly, and the position order of the fluctuation points is rearranged to obtain the rhythm adjustment interval.

[0052] The trend trigger recognition module, based on the signal sequence within the rhythm adjustment interval, marks the position of the upper and lower boundary values ​​of blood pressure in each of the three adjacent detection segments, compares the deviation of the measurement points from the boundary in the current sequence one by one, determines whether there are consecutive measurement points that are all outside the boundary, and then performs trend continuity judgment on the change path and interval time between these measurement points to obtain the arterial pressure change trend recognition result.

[0053] The vital sign linkage extraction module selects respiratory vital signs whose change direction is consistent with the trend based on the trend recognition results of arterial pressure change, identifies the position of the starting change point in the sampling sequence, determines whether the continuous segment after the position has a response pattern, and records the coverage of its fluctuation distribution within the time period to obtain the response trajectory of the associated vital signs.

[0054] The intervention task sorting module records the positional order of all responded vital signs in the associated vital sign response trajectory, the time range of the change, and the frequency of repeated changes on the sampling time axis. It extracts the changer, the change time, and the dense occurrence of changes, and generates a vital sign priority calling sorting structure based on this to obtain a list of vital sign intervention execution order.

[0055] The starting point of the fluctuation delay includes reference time positioning, offset distance quantification, and updated sampling starting point; the rhythm adjustment interval includes time range adjustment, fluctuation sequence rearrangement, and change trend matching; the arterial pressure change trend identification results include exceeding the measurement point boundary, continuous deviation segments, and trend path structure; the associated vital sign response trajectory includes the consistent change starting point, response duration interval, and fluctuation period coverage; and the vital sign intervention execution sequence list includes the order of vital sign appearance, change duration, and change frequency statistics.

[0056] Please see Figure 1 and Figure 2 The fluctuation start point positioning module includes:

[0057] The resistance trend difference detection submodule acquires continuous time period tissue resistance records in the arterial ohmogram detection scenario, extracts the resistance change trend of each time period in the same time series, compares the direction markers of resistance change in any two adjacent time periods, identifies the time period positions with inconsistent directions, and obtains the intervals with inconsistent resistance directions.

[0058] In the context of arterial ohmmetry detection, to obtain tissue resistance records over a continuous period, the entire detection time needs to be divided into multiple consecutive time segments. The length of each segment can be set to 1 second or 0.5 seconds, depending on the refresh rate and signal variation characteristics of the detection device. For example, setting each segment to 1 second means that the number of data points in each segment is related to the sampling rate. If the sampling rate is 100Hz, then each segment contains 100 resistance data points. Next, for each time segment, it is extracted from a unified time series to form a resistance value sequence for the corresponding time point. Based on this, the difference between the resistance values ​​of two adjacent time points within each time segment is calculated. By comparing the magnitude of adjacent resistance values, the direction of resistance change is determined. If the resistance at a later time point is greater than that at a previous time point, it is marked as an increase; otherwise, it is marked as a decrease, forming a complete directional marking sequence within that time segment. After marking the direction of all time periods, compare any two adjacent time periods one by one. If the direction marking at the same time point shows an increase in the previous period and a decrease in the next, or vice versa, it is determined to be a direction inconsistency. Record the position of this time point in the original time series. This allows for the gradual identification of locations where the direction of resistance change conflicts across multiple time periods. When this inconsistency occurs at multiple consecutive time points, these time points are placed side by side to form a resistance direction inconsistency interval. For example, if five consecutive time points have opposite directions and are spaced 0.1 seconds apart, the length of this interval is 0.5 seconds. Through this process, key time periods with reverse fluctuations in the signal change trend can be effectively identified.

[0059] The junction offset anomaly detection submodule extracts the corresponding junction position state and radio wave offset state sequence based on the interval of inconsistent resistance direction, compares the junction state change amplitude with the set junction change recognition threshold, filters the time period position of the offset state anomaly, and generates a junction offset abrupt change position set.

[0060] After identifying the intervals with inconsistent resistance directions, the time points within these intervals are expanded to extract the corresponding records of the interface position state and electromagnetic wave offset state in the original data. The interface position state can be obtained from the position sensor data provided by the detection system, while the electromagnetic wave offset state is derived from the time delay information of the high-frequency detection signal propagating in the medium. These two constitute two types of state sequences corresponding to the time points. When processing the interface position state, the absolute value of the positional change between adjacent time points needs to be calculated to obtain the change amplitude between each pair of time points. The interface change recognition threshold set here is generally determined based on tissue surface movement or detection resolution. For example, if the minimum detection displacement of the device is 0.1 mm, the threshold can be set to 0.3 mm, meaning that a significant displacement is considered to have occurred when the interface position change is greater than or equal to 0.3 mm. The positional change at each time point is compared with this threshold. If the change in any pair of adjacent time points reaches or exceeds this value, the point is identified as a sudden change point in the junction state. Further, it is checked whether the radio wave offset state also changes within a short period before and after this time point. The criterion for judging anomalies in the offset state is whether the change in offset time within a short period (e.g., 0.5 seconds) exceeds a certain threshold. This threshold is determined based on the timing resolution of the equipment. For example, if the resolution of the high-frequency detection equipment is 0.05 milliseconds, then the offset anomaly threshold can be set to 0.4 milliseconds. If the change in the offset state corresponding to a certain time point within a set time window exceeds this value, the radio wave offset state is considered abnormal. Time points that exhibit both junction abrupt changes and offset anomalies are selected, ultimately forming a set of junction offset abrupt change locations for subsequent analysis.

[0061] The fluctuation start point positioning submodule locates the time point sequence in the resistor record based on the position index of the concentrated time period of the interface offset abrupt position, identifies the position of the offset state change, and adjusts the time point backward by a specified amount of time as the new sampling start point to obtain the fluctuation delay start time point.

[0062] After obtaining the indexes of all time periods that meet the conditions in the set of abrupt shifts in the interface offset, these time points need to be further mapped back to the time series of the original resistance data to accurately locate the original time point of each abrupt shift event. After obtaining the original time point sequence, each time point is shifted backward by a set amount of time as a new sampling starting point. The setting of this time delay needs to be determined based on the typical delay characteristics of the tissue response and the system signal processing cycle, and can generally be set between 1.0 and 2.0 seconds. For example, combined with observations of multiple samples, it was found that the signal disturbance response delay is generally around 1.2 seconds, so the delay time can be set to 1.2 seconds to ensure that subsequent sampling covers the complete change process of the disturbance response. For each abrupt shift time point, the sampling starting point is recalibrated by 1.2 seconds, forming a new set of fluctuation starting time points. The new set of time points will serve as a key entry point in the subsequent resistance signal characteristic analysis to extract the fluctuation response change trend after the disturbance occurs, assisting in further state identification and structural stability assessment.

[0063] Please see Figure 1 and Figure 2 The rhythm structure adjustment module includes:

[0064] The direction coherence recognition submodule extracts the sampling sequence within the current fluctuation segment based on the starting position of the fluctuation delay start point, records the change direction of the fluctuation range corresponding to the time point, extracts the change direction marker within the previous segment, compares the consistency between the current segment and the previous segment in the corresponding time direction, filters out continuous time periods with consistent direction, and obtains the direction coherence interval.

[0065] Based on the starting position of the fluctuation delay start point, firstly, according to each calibrated start time point, continuous data of a certain duration is extracted from the original resistance signal to form a sampling sequence within the current fluctuation segment. This duration can be set to 2 seconds or 3 seconds. Combined with the actual sampling frequency, a corresponding number of data points are obtained. For example, if the sampling rate is 100Hz, the number of sampling points per segment is 200 or 300. Then, the adjacent difference is calculated for each data point in the sampling sequence to determine the magnitude relationship between each pair of adjacent data points. If the value of the later data point is greater than the value of the previous data point, it is recorded as "rising"; otherwise, it is recorded as "falling," forming a complete fluctuation direction sequence. Finally, the fluctuation direction is summarized sequentially for all data points in the current fluctuation segment. This process constructs a list of change directions for the current segment. Simultaneously, it retrieves the previous segment adjacent to the current segment, extracts the change direction markers from that segment, and compares the direction values ​​at each corresponding position on the time axis between the two segments. If the direction of a point in the current segment is consistent with the direction of the same position in the previous segment, then that point is considered a point of consistent direction. This point is then extended forward and backward point by point to find the largest continuous consistent region. For example, if the direction markers of the 10th to 30th data points in the current segment are exactly the same as those of the 10th to 30th points in the previous segment, then it is considered a continuous directional interval of 21 data points. This operation will completely filter out the continuous time periods in the fluctuation segment where the direction is consistent and record their time range, thus obtaining the continuous directional interval.

[0066] The segment boundary revision submodule identifies the degree of overlap between the start and end times of the current fluctuation segment and the coverage area based on the time boundary of the directional continuity interval, and determines whether it meets the conditions for extension and compression. If it does, the start and end times of the current fluctuation segment are revised to generate the adjusted segment range.

[0067] Based on the time boundaries of the directional continuity intervals obtained above, the start and end times are first recorded respectively. The overall time range of the current fluctuation segment is then compared, i.e., the time interval formed by the original start and end points. The time overlap range and coincidence ratio between the start and end times of the directional continuity interval and the start and end times of the original segment are calculated. If the start time of the continuity interval is earlier than the original start time by more than a set extension threshold, or the end time is later than the original end time by more than a set compression threshold, then boundary revision is performed. The extension and compression thresholds are set based on historical fluctuation segment stability statistics. For example, the extension threshold is set to 0.8 seconds and the compression threshold to 0.6 seconds. If the start time of the directional continuity interval is 1.0 second earlier than the original start time, the extension condition is met, and the start time point is revised forward. If the end time is 0.7 seconds later than the original end time, the compression condition is met, and the end time point is revised backward. This completes the redefinition of the start and end boundaries, resulting in a new time segment range as the adjusted segment range of the current fluctuation segment.

[0068] The rhythm reordering submodule calls the revised time range in the adjusted segment range, reorders the contained fluctuation points according to the original fluctuation direction, and uses the sorted fluctuation point position sequence as the segment benchmark to obtain the rhythm adjustment interval.

[0069] The revised time range within the adjusted fluctuation range is retrieved, and the position indices and corresponding fluctuation direction markers of all fluctuation points within that time period are re-extracted. A preliminary sorting is performed based on the direction markers. First, all points marked "rising" are extracted and arranged in chronological order, then all points marked "falling" are extracted and arranged in chronological order, forming a fluctuation point sequence prioritizing direction consistency. For example, if a segment contains 80 data points after adjustment, with 45 points in the "rising" direction and 35 points in the "falling" direction, the two subsets are sorted by time and concatenated to form a new sorted sequence. This sequence is then used to reconstruct the segment baseline, serving as the interval result after rhythm adjustment. The baseline arrangement references the changing trend of the original fluctuation direction to ensure that the sorted sequence reflects the main trend characteristics of the fluctuation rhythm, ultimately generating the rhythm adjustment interval.

[0070] Please see Figure 1 and Figure 2 The trend trigger recognition module includes:

[0071] The boundary extraction and calibration submodule extracts the upper and lower boundaries of blood pressure in three adjacent detection segments based on the signal sequence within the rhythm adjustment interval, calibrates the time node of the boundary value in each segment, records the index sequence of the boundary time point in the segment, and generates a boundary position index set.

[0072] Based on the signal sequence within the rhythm adjustment interval, the signal sequence is first segmented according to the adjusted time boundaries. Three consecutive detection segments are extracted, and the blood pressure values ​​of all sampling points in each segment are labeled. The maximum value in each segment is identified as the upper boundary of blood pressure for that segment, and the minimum value as the lower boundary. Furthermore, the actual location of each boundary value in the time series is determined, and its sampling index number within the segment is recorded. For example, if the sampling start point of the first segment is 0, and the upper boundary value appears at the 42nd sampling point, its index is 4. 2. Store the upper and lower boundary position indices of all segments one by one to form a complete set of boundary position indexes. If the sampling frequency is 100Hz and each segment is 2 seconds, then each segment contains 200 data points. Let the upper and lower boundary indices of the three segments be 45 and 10 for the first segment, 49 and 8 for the second segment, and 51 and 12 for the third segment. Then the boundary index set is {(45, 10), (49, 8), (51, 12)}. This index set will serve as the basis for subsequent boundary judgment and trend tracking.

[0073] The out-of-boundary interval filtering submodule calls the corresponding positions of the upper and lower limits of the boundary in the boundary position index set, extracts the position values ​​of the measurement points in the same segment, compares the positional relationship between each measurement point and the corresponding boundary value, and filters the time period positions where consecutive measurement points exceed the upper and lower boundaries to obtain the continuous out-of-boundary point intervals.

[0074] The system calls upon the upper and lower boundary positions stored in the aforementioned boundary position index set, extracts the corresponding blood pressure values ​​as the judgment benchmark, and then iterates through the position values ​​of all sampling points within the same segment, judging whether the value exceeds the upper boundary or is lower than the lower boundary point. If a measurement point value is greater than the upper boundary value of the corresponding segment or less than the lower boundary value, the point is marked as an out-of-bounds point. All consecutive out-of-bounds points are recorded to form an out-of-bounds point time index sequence. Further analysis is conducted to determine whether the out-of-bounds points are consecutive time points. For example, if the blood pressure value of each sampling point from the 45th to the 60th point exceeds the upper boundary of the segment by 150 mmHg, then these 16 sampling points constitute a consecutive out-of-bounds interval. Or, if the blood pressure value of all sampling points from the 5th to the 20th point is lower than the lower boundary by 90 mmHg, it also constitutes a consecutive out-of-bounds interval. All out-of-bounds point time periods that meet the "continuity" condition are selected, where the "continuity" condition is that the index difference between two adjacent out-of-bounds points is 1. If there is a gap, it is not included in the consecutive interval. Finally, all segments that meet this standard are retained as consecutive out-of-bounds point intervals.

[0075] The trend consistency determination submodule extracts the change path direction markers between measuring points and the time interval sequence between adjacent measuring points based on the position information of measuring points in the continuous boundary point interval, determines whether the change direction is consistent and the interval does not exceed the trend continuity threshold, and obtains the arterial pressure change trend recognition result.

[0076] Based on the location information of all measuring points within the continuous boundary crossing interval, the direction of blood pressure change between each pair of adjacent measuring points is first extracted, i.e., whether the value of the later measuring point is higher or lower than the value of the previous point, determining whether its direction is "rising" or "falling," forming a corresponding change path direction sequence. Simultaneously, the sampling time interval between each pair of adjacent points is statistically analyzed. If the sampling frequency is 100Hz, the interval between each sampling point is 0.01 seconds. The time interval between each pair of measuring points is compared with a set trend continuity threshold, set here to 0.02 seconds, derived from the device's minimum cycle statistics for fluctuation continuity response, for example, by analyzing 10... The trend duration of 0 samples was statistically analyzed, and it was found that the trend was continuous when the time between measurement points was less than 0.02 seconds and the direction was consistent. Therefore, this value was selected as the threshold benchmark. If the change direction of consecutive point pairs is consistent and the corresponding time interval is less than or equal to 0.02 seconds, it is considered that the trend continues. If the change direction between a pair of points is inconsistent or the time interval is greater than the threshold, the trend is interrupted. The continuous measurement point segments that meet the conditions are marked as the same trend block. After statistical analysis of all trend blocks, it can be determined whether there is a significant upward or downward trend in arterial pressure in the entire continuous out-of-bounds interval, and finally the arterial pressure change trend identification result is obtained.

[0077] Please see Figure 1 and Figure 2 The vital sign linkage extraction module includes:

[0078] The direction matching and filtering submodule extracts the change direction markers of the sign sequences in respiratory-related signs based on the trend information in the arterial pressure change trend recognition results. It then matches and filters the change direction of each sign with the trend direction, retaining the sign data sequences with consistent directions to obtain the direction-matched signs.

[0079] Based on the trend information from the arterial pressure change trend identification results, the directional markers clearly indicated in the trend are first read. For example, upward indicates a continuous increase in arterial pressure, downward indicates a continuous decrease, or unchanged indicates a stable state. Then, respiratory-related vital sign data corresponding to this time period are extracted. Common vital sign sequences include respiratory rate, tidal volume, and chest and abdominal movement amplitude. Each vital sign is constructed into a sequence based on continuous sampling results in chronological order. The difference between two adjacent sampling points within each sequence is calculated to determine whether it is increasing, decreasing, or stable, thus generating a complete sequence of change direction markers for each vital sign, such as chest and abdominal movement amplitude. If the sequence is {2.1, 2.3, 2.5, 2.2}, then the direction sequence is {rising, rising, falling}. The direction sequence of each vital sign is then compared one by one with the direction marker of the arterial pressure trend. If a segment of a vital sign shows a continuous change in direction that perfectly matches the arterial pressure trend, then the vital sign data within that segment is marked as a direction-matched vital sign. For example, if the arterial pressure is rising, and the chest and abdominal movements are all rising in the direction from the 10th to the 40th sampling point, then that segment is retained. This process is repeated to match and filter all vital signs, and finally, the vital sign data sequences with consistent directions are retained and summarized to form direction-matched vital signs.

[0080] The morphological response segment extraction submodule calls the direction matching vital signs, identifies the location index of the direction change, records the position number of the starting point of the vital sign change in the original sampling sequence, determines whether there is a change trajectory in the continuous time period after the number that exceeds the morphological recognition benchmark value, and obtains the morphological response segment.

[0081] The selected directional matching vital signs are used to detect the first time point of directional change in each vital sign change sequence. The position number of this change point in the original vital sign sampling sequence is recorded. For example, if the original sequence is {rise, rise, rise, fall, fall}, then the first directional change occurs at the 4th sampling point. The difference of vital sign values ​​in the continuous time period after this number is calculated to extract the fluctuation amplitude. It is determined whether the difference between two adjacent sampling points exceeds the set morphological recognition benchmark value. This benchmark value is set based on the accuracy of the vital sign measurement equipment and the normal human body fluctuation range. The standard is set to 0.15 units. For example, the recognition benchmark for chest and abdominal movement amplitude is set to 0.15cm. If multiple differences in the continuous difference sequence are greater than or equal to 0.15cm, it indicates that there is obvious morphological fluctuation in this time period. The segment that continuously meets this condition is located as the morphological response segment, and its start and end time point index is recorded to ensure that the morphological response segment has substantial fluctuation characteristics and trend change response capability.

[0082] The response trajectory generation submodule extracts the covered time interval based on the start and end time index of the morphological response segment, calculates the fluctuation frequency value within the response segment, marks the distribution structure of the fluctuation interval within the time interval, and obtains the related vital sign response trajectory.

[0083] The specific formula for calculating the fluctuation frequency value within the response range is as follows:

[0084] ;

[0085] in, Representing the Fluctuation frequency value within each response segment Representing the The number of time index points within each response segment Representing the Within the first response segment The difference between the vital sign value at each time point and the vital sign value at the previous time point, i.e. , Representing the The standard deviation of phenotypic values ​​within each response segment Representing the The average value of the symptom within each response segment Representing the The maximum value of the eigenvalue within each response segment Representing the The minimum value of the eigenvalue within each response segment Representing the Within the first response segment Vital value at each time point;

[0086] The formula employs various mathematical operations, including summation (∑), subtraction (-), division ( / ), and absolute value (||). The summation symbol represents a calculation process repeated at multiple time points throughout the entire segment; starting from the second point, the difference needs to be calculated based on the previous time point. The absolute value operation ensures that the direction of change does not affect the intensity of the fluctuation. The difference term is as follows: It measures the stationarity or continuity of fluctuations; standard deviation normalization and maximum / minimum value normalization improve dimensional consistency.

[0087] Parameter calculation and numerical examples:

[0088] Instructions for selecting time intervals: Let the first time interval be... Each response segment is arrive The recording frequency is once per second, for a total of Data points.

[0089] Vital signs (Unit: mmHg): Obtained by a continuous blood pressure monitor, for example:

[0090] , , , , , .

[0091] Change value ;

[0092] For example:

[0093] ;

[0094] ;

[0095] ;

[0096] ;

[0097] .

[0098] mean with standard deviation :

[0099] The formula for calculating the mean is:

[0100] ;

[0101] Assumption There are a total of 61 values, calculated as follows: mmHg;

[0102] The formula for calculating standard deviation is:

[0103] ;

[0104] have to mmHg.

[0105] Maximum value and minimum value :

[0106] Maximum vital signs in the monitoring segment mmHg, minimum value mmHg.

[0107] Example calculations using the formula (calculating with the first 5 sets of data points) are shown in Table 1 below:

[0108] Table 1: Monitoring Parameters for Vital Sign Response Zones

[0109]

[0110] As shown in Table 1, the parameters required for the calculation have been quantified and listed.

[0111] Calculation of normalized fluctuation frequency value (First 5 items):

[0112] ;

[0113] Right now

[0114] ;

[0115] ;

[0116] The result indicates that the fluctuation frequency value within the response range The value was significantly higher than the baseline value of 20 (the baseline value was calculated by taking the average value of previous static measurements of vital signs and setting the normal range of variation), indicating that the vital signs fluctuated frequently and drastically during this time period, which requires close attention.

[0117] Advantages of the formula:

[0118] By introducing terms for absolute variation of vital sign differences, rate of change of differences, and standardized bias, the overall effect of vital sign fluctuations, frequency, and outlier characteristics over time series is achieved, thus making the results more comprehensive. It also possesses amplitude sensitivity, trend recognition capability, and abnormal response capture capability, enhancing the reliability of vital sign response trajectory discrimination and the accuracy of time positioning.

[0119] Explanation of the normalization method for each parameter:

[0120] ;

[0121] Objective: To convert the variation of vital signs between adjacent time points into relative standard deviation units in order to eliminate the influence of the absolute numerical scale of the vital signs themselves;

[0122] Method: Divide the change value (unit: e.g. mmHg) by the standard deviation of the corresponding segment (unit: mmHg), and the result is a dimensionless value;

[0123] Normalization logic: By standardizing the instantaneous changes in vital signs according to the standard deviation, we can measure whether the changes are significant.

[0124] ;

[0125] Objective: To reflect the degree of abrupt change in the rate of change of vital signs;

[0126] Method: This item is the difference between two adjacent changes, with the unit remaining unchanged (e.g., mmHg), and is not included in normalization;

[0127] Normalization: This term is no longer standardized in the formula, preserving its original numerical differences to enhance the ability to respond to drastic changes.

[0128] ;

[0129] Objective: To measure the relative extent to which the current value of a vital sign deviates from the mean of its segment, and to remove the influence of the absolute dimensions of the upper and lower limits of the segment;

[0130] Method: Divide the deviation of each point's vital sign value from the mean by the range of that segment (maximum value minus minimum value), with the unit being mmHg. The result is dimensionless.

[0131] Normalization logic: The larger this ratio is, the more the current vital signs deviate from the normal fluctuation range.

[0132] Definition of fluctuation frequency value within the response range:

[0133] Fluctuation frequency value It refers to the cumulative intensity of amplitude and rate fluctuations in vital sign values ​​over time within a specific response range. It not only reflects the instantaneous changes in a single point of vital sign value, but also comprehensively considers the continuity and trend of the changes.

[0134] The fluctuation frequency value differs from a simple count of the number of changes. Instead, it combines the magnitude of each change (standardized), the continuity of fluctuation (through the difference), and the degree of deviation from the central trend (normalized deviation) in a weighted manner, and sums them over the entire segment to obtain a dimensionless intensity index.

[0135] Explanation of the operation principle:

[0136] Starting conditions: Starting from the second time point within the segment, for each time point... Perform fluctuation calculations;

[0137] First item : Used to capture whether the magnitude of the change in vital signs at this moment is statistically abnormal. A value greater than 1 usually indicates that the change at this point exceeds one standard deviation;

[0138] Second item This reflects the continuity and stability of the process of changes in vital signs; the larger the value, the more unstable the process of change.

[0139] Third item : Measure the relative position of the vital sign value at that moment within the overall segment. If it deviates significantly from the median, it indicates that there is a significant abnormality at that point.

[0140] Finally, by summing the above three items at all time points throughout the entire segment, we obtain... It is used to quantitatively characterize the frequency and total intensity of vital signs fluctuations within a given time period.

[0141] This value can serve as a basic indicator for subsequent tasks such as identifying abnormal response regions, feature clustering, and labeling vital signs trends. The larger the value, the more frequent and drastic the fluctuations in vital signs within that response segment.

[0142] Please see Figure 1 and Figure 2 The intervention task ranking module includes:

[0143] The vital sign first-order sorting submodule extracts the first response position of the vital sign on the sampling time axis based on all the responded vital signs in the associated vital sign response trajectory, numbers and sorts the first and first response points of the differentiated vital signs in chronological order, records the time index arrangement structure of each vital sign, and generates the vital sign time series order value.

[0144] Based on all responded vital signs in the associated vital sign response trajectory, the earliest response point is first extracted from the response trajectory of each vital sign. The starting response time index of each vital sign on the sampling time axis is read sequentially, and this time index is recorded as the first response point of that vital sign. For example, if the fluctuation of vital sign A first exceeds the threshold at the 102nd sampling point, vital sign B first responds at the 89th sampling point, and vital sign C first responds at the 95th sampling point, then the corresponding time indices are 102, 89, and 95, respectively. All the first response points of vital signs are sorted in ascending order of time index to obtain the order sequence: vital sign B, vital sign C, and vital sign A, and assigned sequential numbers 1, 2, and 3, respectively. The correspondence between the sorted vital sign names and their time indices is recorded to form a vital sign time sequence order value table. This table completely records the chronological order information of vital sign responses on the time axis, which constitutes the key basis for subsequent intervention sorting.

[0145] The vital sign duration segment calculation submodule calls the time index of the vital sign in the sequential value of the vital sign time series, extracts the start and end boundaries of the continuous change of the vital sign in the sampling sequence, calculates the duration value of the change, and marks the complete segment index during the change period to obtain the duration interval of the vital sign change.

[0146] The specific formula for calculating the duration of change is as follows:

[0147] ;

[0148] in, Representative number is physical signs The duration of change during a certain continuous change process. Indicating physical signs The Index of the end time point of each change segment. Indicating physical signs The Index of the starting time point of each change segment. Indicating physical signs In the Within the first change segment The difference in vital sign values ​​between each sampling point and its previous sampling point Indicating physical signs In the Within the first change segment The time difference between each sampling point and its previous sampling point Indicating physical signs In the The average rate of change of all sampling point pairs within each change segment. Indicating physical signs In the Number of sampling point pairs within each variation segment Indexed by the difference sequence number;

[0149] Parameter acquisition and practical examples:

[0150] Calculate the average rate of change:

[0151] ;

[0152] Calculate the term for each pair of sampling points:

[0153] Pair 1:

[0154] ;

[0155] Pair 2:

[0156] ;

[0157] The third pair:

[0158] ;

[0159] 4th pair:

[0160] ;

[0161] Calculate the average:

[0162] ;

[0163] Calculate the final duration of the change:

[0164] ;

[0165] Results analysis:

[0166] The results indicate that the duration of the change in vital sign α\alphaα was 4.846 minutes during the 10-minute to 14-minute interval.

[0167] The advantages of the formula are:

[0168] By introducing the square root of the product of the difference in vital signs and the time difference, we can more sensitively capture the magnitude and speed of changes in vital signs; and by subtracting the term of the difference in vital signs divided by the average rate of change, we can adjust for deviations caused by abnormal fluctuations, thereby more accurately reflecting the actual duration of changes in vital signs.

[0169] Explanation of the dimension normalization method for each parameter in the formula:

[0170] To ensure comparability, merging, and computational stability among data with different dimensions, all parameters involved in the formula need to be normalized. This is achieved by combining standard normalization (Z-score normalization) and min-max normalization. The normalization methods for each parameter are as follows:

[0171] Time index parameters ( , , );

[0172] This is used to ensure that time values ​​are distributed between 0 and 1, improving the alignment between multiple time data segments.

[0173] Vital sign difference parameter ( );

[0174] in This represents the mean of the differences in physical characteristics among the samples. Its standard deviation; this method is used to eliminate the influence between different signs scales.

[0175] Average rate of change parameter ( );

[0176] in , These represent the minimum and maximum rates of change of this trait in historical data.

[0177] When square root, product, and ratio terms appear in mixed expressions, their results need to be mapped to a unified standard interval based on the normalization factor (e.g., setting the maximum absolute value of all vital sign changes as a benchmark). Or centralize to 0.

[0178] Definition of duration of change:

[0179] The duration of change refers to the actual length of time during which a specific vital sign undergoes a significant change in its trend or value within a continuous interval in a monitored vital sign time series. This time not only considers the start and end times of the change interval but also incorporates factors such as the intensity, speed, and amplitude of the change, thereby comprehensively quantifying the nature of "continuous change."

[0180] This value is an important indicator reflecting the stability and trend continuity of changes in vital signs, and is applicable to various scenarios such as medical monitoring (e.g., blood pressure, heart rate) and industrial process control (e.g., temperature changes).

[0181] Explanation of the operational principle (textual explanation):

[0182] The calculation of the duration of this change is based on two core components:

[0183] Measurement of the base duration period:

[0184] The basic duration of change is obtained by measuring the difference between the end and start times of a period of change in vital signs. This part directly quantifies the absolute length of the change period and is a fundamental dimension for analyzing the persistence of change.

[0185] Weighted adjustment of dynamic factors of vital signs changes:

[0186] For each pair of sampling points, calculate the absolute value of the product of the difference in vital signs and the difference in time, and take its square root to measure the intensity of instantaneous changes.

[0187] At the same time, the ratio of the difference in vital signs between each pair of sampling points to the average rate of change in that segment is calculated as a correction term to suppress violent abnormal fluctuations;

[0188] The difference between the two reflects the "real change effect" of the current sampling point pair. Then, the summation and averaging of all sampling point pairs are used to obtain the change dynamic correction term.

[0189] Add the correction term to the base duration and take the absolute value of the entire result to avoid negative values ​​caused by reversing the start and end order.

[0190] The final value for the duration of change is an integrated expression of the three-dimensional characteristics of the change range, intensity, and speed, and its composition reflects the following logical structure:

[0191] "Time span + average adjustment of change range" and "composite calculation" are achieved by using multiple mathematical operations such as multiplication, square root, ratio and absolute value to achieve data-driven change recognition and duration estimation capabilities.

[0192] The comprehensive sorting and list generation submodule calculates the frequency of each vital sign change in the entire sequence based on the index of the change segment of vital signs in the duration interval of vital sign changes. It uses time order, duration, and frequency of occurrence as sorting criteria, establishes a multi-factor combination judgment logic, and obtains a list of vital sign intervention execution order.

[0193] Based on the index information recorded for each segment within the duration interval of vital sign changes, the number of changes in each vital sign throughout the entire sampling sequence is counted, i.e., the number of response segments for the same vital sign. For example, vital sign A appears 3 times in the entire signal segment, vital sign B appears 5 times, and vital sign C appears 2 times, with their frequencies recorded as 3, 5, and 2 respectively. The time sequence number, cumulative duration, and frequency of each vital sign are extracted as comprehensive evaluation factors, and a judgment logic is constructed to sort them sequentially. The priority weights in the judgment logic are set as follows: time sequence factor 0.5, duration factor 0.3, and frequency factor 0.2. The ranking value of each vital sign is calculated according to the weights of each factor. For example, vital sign B has a time sequence of 1 (value 1), a total duration of 2.5 seconds, and a frequency of 5 times, while vital sign C has a time sequence of 2 (value 2), a total duration of 1.2 seconds, and a frequency of 2 times. The ranking values ​​are calculated and compared according to the rules, and finally, the ranking values ​​are arranged from smallest to largest to generate a list of vital sign intervention execution sequences.

[0194] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A health management and physical examination big data platform, characterized in that, include: The fluctuation start point localization module acquires tissue resistance records for a time period in the arterial ohmogram detection scenario, determines the direction of adjacent resistance changes, compares the difference between the change position of the interface and the electrical wave offset, selects the deviation time point, and moves it backward as the new sampling start point to obtain the fluctuation delay start time point; The rhythm structure adjustment module sets a starting point based on the start time of the fluctuation delay, compares the current and previous fluctuation directions, and if they are consistent, adjusts the segment length and rearranges the order of fluctuation points to obtain the rhythm adjustment interval. The trend trigger recognition module, based on the rhythm adjustment interval, calibrates the blood pressure boundary position, determines whether the measuring point continuously falls outside the boundary, and makes a continuous judgment on the change path and interval to obtain the arterial pressure change trend recognition result. The vital sign linkage extraction module, based on the trend direction in the arterial pressure change trend recognition result, selects items that are consistent with the changes in expiratory time, inspiratory holding time and lung volume, compares the initial response position and the coverage of the continuous segment, and obtains the associated vital sign response trajectory. The fluctuation delay start point includes reference time positioning, offset distance quantification, and updated sampling start point; the rhythm adjustment interval includes time range adjustment, fluctuation sequence rearrangement, and change trend matching; the arterial pressure change trend identification result includes exceeding the measurement point boundary, continuous deviation segment, and trend path structure; the associated vital sign response trajectory includes consistent change start point, response duration interval, and fluctuation period coverage. The fluctuation start point location module includes: The resistance trend difference detection submodule acquires continuous time period tissue resistance records in the arterial ohmogram detection scenario, extracts the resistance change trend of each time period in the same time series, compares the direction markers of resistance change in any two adjacent time periods, identifies the time period positions with inconsistent directions, and obtains the intervals with inconsistent resistance directions. The junction offset anomaly detection submodule extracts the corresponding junction position state and radio wave offset state sequence based on the interval of inconsistent resistance direction, compares the junction state change amplitude with the set junction change recognition threshold, filters the time period position of the offset state anomaly, and generates a junction offset abrupt change position set. The fluctuation start point positioning submodule, based on the position index of the time period of the concentrated time of the interface offset abrupt position, locates the time point sequence in the resistor record, identifies the position of the offset state change, and adjusts the time point backward by a specified amount of time as the new sampling start point to obtain the fluctuation delay start time point; The trend trigger recognition module includes: The boundary extraction and calibration submodule extracts the upper and lower blood pressure boundaries of three adjacent detection segments based on the signal sequence within the rhythm adjustment interval, calibrates the time nodes of the boundary values ​​in each segment, records the index sequence of the boundary time points in the segments, and generates a boundary position index set. The boundary crossing interval filtering submodule calls the corresponding positions of the upper and lower limits of the boundary in the boundary position index set, extracts the position values ​​of the measurement points in the same segment, compares the positional relationship between each measurement point and the corresponding boundary value, and filters the time period positions where consecutive measurement points exceed the upper and lower boundaries to obtain the continuous boundary crossing point intervals. The trend consistency determination submodule extracts the change path direction markers between measuring points and the time interval sequence between adjacent measuring points based on the position information of the measuring points in the continuous boundary point interval, determines whether the change direction is consistent and the interval does not exceed the trend continuity threshold, and obtains the arterial pressure change trend recognition result.

2. The health management and physical examination big data platform according to claim 1, characterized in that, The rhythm structure adjustment module includes: The direction coherence recognition submodule extracts the sampling sequence within the current fluctuation segment based on the starting position of the fluctuation delay start point, records the change direction of the fluctuation range corresponding to the time point, extracts the change direction marker within the previous segment, compares the consistency between the current segment and the previous segment in the corresponding time direction, filters out continuous time periods with consistent direction, and obtains the direction coherence interval. The segment boundary revision submodule identifies the degree of overlap between the start and end times of the current fluctuation segment and the coverage area based on the time boundary of the directional continuous interval, and determines whether it meets the conditions for extension and compression. If it does, the start and end times of the current fluctuation segment are revised to generate the adjusted segment range. The rhythm reordering submodule calls the revised time range in the adjusted segment range, reorders the contained fluctuation points according to the original fluctuation direction, and arranges the sorted fluctuation point position sequence as the segment benchmark to obtain the rhythm adjustment interval.

3. The health management and physical examination big data platform according to claim 1, characterized in that, The vital sign linkage extraction module includes: The direction matching and filtering submodule extracts the change direction markers of the sign sequences in the respiratory-related signs based on the trend information in the arterial pressure change trend recognition results. It then matches and filters the change direction of each sign with the trend direction, retains the sign data sequences with consistent directions, and obtains the direction-matched signs. The morphological response segment extraction submodule calls the direction matching vital signs, identifies the location index of the direction change, records the position number of the starting point of the vital sign change in the original sampling sequence, determines whether there is a change trajectory in the continuous time period after the number that exceeds the morphological recognition benchmark value, and obtains the morphological response segment. The response trajectory generation submodule extracts the covered time interval based on the start and end time index of the morphological response segment, calculates the fluctuation frequency value within the response segment, marks the distribution structure of the fluctuation interval within the time interval, and obtains the associated vital sign response trajectory.

4. The health management and physical examination big data platform according to claim 3, characterized in that, The specific formula for calculating the fluctuation frequency value within the response range is as follows: ； in, Representing the Fluctuation frequency value within each response segment Representing the The number of time index points within each response segment Representing the Within the first response segment The difference between the vital sign value at each time point and the vital sign value at the previous time point, i.e. , Representing the The standard deviation of phenotypic values ​​within each response segment Representing the The average value of the symptom within each response segment Representing the The maximum value of the eigenvalue within each response segment Representing the The minimum value of the eigenvalue within each response segment Representing the Within the first response segment Vital signs at each time point.

5. The health management and physical examination big data platform according to claim 1, characterized in that, It also includes an intervention task ranking module: The intervention task sorting module identifies changes, durations, and frequencies of occurrence of vital signs in the associated vital sign response trajectory, and combines the intervention order to obtain a list of vital sign intervention execution order. The list of intervention sequence for vital signs includes the order of appearance of vital signs, duration of changes, and frequency of changes.

6. The health management and physical examination big data platform according to claim 5, characterized in that, The intervention task sorting module includes: The vital sign first-order sorting submodule extracts the first response position of the vital sign on the sampling time axis based on all the responded vital signs in the associated vital sign response trajectory, numbers and sorts the first and first response points of the differentiated vital signs in chronological order, records the time index arrangement structure of each vital sign, and generates the vital sign time series order value. The vital sign duration segment calculation submodule calls the time index of the vital sign in the sequential value of the vital sign time series, extracts the start and end boundaries of the continuous change of the vital sign in the sampling sequence, calculates the duration value of the change, and marks the complete segment index during the change period to obtain the duration interval of the vital sign change. The integrated sorting and list generation submodule calculates the frequency of each vital sign change in the entire sequence based on the index of the change segment of vital signs in the duration interval of the vital sign change. It then uses time order, duration, and frequency of occurrence as sorting criteria to establish a multi-factor combination judgment logic and obtain a list of vital sign intervention execution order.

7. The health management and physical examination big data platform according to claim 6, characterized in that, The formula for calculating the duration of the change is as follows: ； in, Representative number is physical signs The duration of change during a certain continuous change process. Indicating physical signs The Index of the end time point of each change segment. Indicating physical signs The Index of the starting time point of each change segment. Indicating physical signs In the Within the first change segment The difference in vital sign values ​​between each sampling point and its previous sampling point Indicating physical signs In the Within the first change segment The time difference between each sampling point and its previous sampling point Indicating physical signs In the The average rate of change of all sampling point pairs within each change segment. Indicating physical signs In the Number of sampling point pairs within each variation segment This is the index for the difference sequence number.

Images