A quality assessment method and system based on geriatric syndrome data

Abstract

This invention belongs to the field of data analysis technology, specifically relating to a quality assessment method and system based on geriatric syndrome data. The method identifies data collection nodes in geriatric syndrome data according to preset anomaly judgment criteria and marks them as checkpoints. When data fluctuation exceeds an anomaly threshold and the difference exceeds a difference threshold, the corresponding data collection node is marked as a checkpoint. This allows for the screening of data collection nodes caused by sudden state changes from continuous geriatric syndrome data, simultaneously focusing on the absolute magnitude of data fluctuation and its rate of change over time. This effectively avoids misjudgments caused by regular data fluctuations, improving the accuracy and relevance of the quality assessment starting point and checkpoint location.

Description

Technical Field

[0001] This invention belongs to the field of data analysis technology, specifically relating to a quality assessment method and system based on geriatric syndrome data. Background Technology

[0002] Geriatric syndromes are diseases involving the decline of multiple physiological systems. Effective monitoring and early warning of these syndromes are of great significance for improving the quality of life of older adults and preventing adverse health events. To achieve proactive health management and risk intervention, it is necessary to utilize modern information technology for continuous intelligent analysis of physiological data related to geriatric syndromes.

[0003] Current data assessment technologies for geriatric syndromes employ batch processing or centralized computing models after data accumulation. This means that a large amount of data is collected and then analyzed uniformly. This results in the system's inability to respond quickly to instantaneous changes in a patient's health condition, missing crucial intervention windows. In the event of acute deterioration, it fails to provide timely warnings. Because the various indicators of geriatric syndromes are complex and interconnected, existing technologies often rely on simple threshold judgments or fluctuation analysis of single data curves. This makes it difficult to effectively distinguish between normal physiological fluctuations and pathological precursors, leading to increased false alarm and false negative rates.

[0004] To address the aforementioned problems, this invention provides a quality assessment method and system based on geriatric syndrome data, thereby resolving the technical issues existing in the prior art. Summary of the Invention

[0005] The purpose of this invention is to provide a quality assessment method and system based on geriatric syndrome data to solve the problem of untimely assessment of geriatric syndrome data.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A quality assessment method based on geriatric syndrome data involves performing a quality assessment when data collection nodes meeting preset anomaly criteria are detected within a preset main time period.

[0008] The quality assessment includes:

[0009] Based on the acquisition nodes that meet the preset anomaly judgment criteria, they are marked as checkpoints and the offset time interval is determined;

[0010] Within the offset time interval, at least one node to be evaluated is determined, and the collected data corresponding to the node to be evaluated is determined as the data to be evaluated.

[0011] And score the data to be evaluated according to the preset scoring criteria to generate an overall evaluation result, and output an evaluation report based on the overall evaluation result;

[0012] Among them, those that meet the preset anomaly judgment criteria include:

[0013] After dividing the preset main time period into multiple time segments, if the data fluctuation amplitude of a certain time segment exceeds the preset abnormal threshold, and the difference between the data fluctuation amplitude of that time segment and the data fluctuation amplitude of its adjacent time segments exceeds the preset difference threshold;

[0014] Determining at least one node to be evaluated within the offset time interval includes:

[0015] Multiple nodes to be evaluated are determined based on preset node selection rules.

[0016] Preferably, calibrating as a checkpoint and determining the offset time interval includes:

[0017] Based on the time of the checkpoint, the offset time interval is defined by the set forward offset time and backward offset time.

[0018] Preferably, scoring the data to be evaluated according to preset scoring criteria includes:

[0019] An initial score is assigned to the data to be evaluated corresponding to each node based on the time distance between each node to be evaluated and the checkpoint.

[0020] In addition, the initial scores are adjusted according to the preset weight adjustment rules to obtain the scores of the data to be evaluated.

[0021] Preferably, the method further includes:

[0022] After scoring the data to be evaluated, the node corresponding to the data with the highest score is determined as the reference node, and its data is determined as the benchmark. The difference between the other data to be evaluated and the benchmark is calculated. When the difference exceeds the preset fluctuation tolerance threshold, the node corresponding to the difference is marked as a checkpoint to trigger the quality assessment.

[0023] Preferably, the data to be evaluated is scored according to a preset scoring standard to generate an overall evaluation result, including:

[0024] Statistical indicators are generated based on the average score of all data to be evaluated and the total number of nodes to be evaluated.

[0025] In addition, the statistical indicators are compared with preset reference standards, and a level-one early warning message is generated in the evaluation report when the statistical indicators are lower than the reference standards.

[0026] This invention also discloses a quality assessment system based on geriatric syndrome data, comprising:

[0027] The anomaly detection module is used to acquire geriatric syndrome data collected within a preset main time period and monitor the data to identify collection nodes that meet preset anomaly judgment criteria, which are then used as evaluation trigger events.

[0028] The quality assessment module is used to perform quality assessments for assessment trigger events identified by the anomaly detection module. The quality assessment module is configured as follows:

[0029] The acquisition nodes that meet the preset anomaly judgment criteria are marked as checkpoints, and the offset time interval is determined based on the checkpoints;

[0030] Within the offset time interval, the node to be evaluated is determined, and the data to be evaluated corresponding to the node to be evaluated is scored according to the preset scoring criteria.

[0031] After scoring the data to be evaluated, the node corresponding to the data with the highest score is determined as the reference node, and other data to be evaluated are compared with the benchmark corresponding to the reference node. When the comparison result exceeds the preset fluctuation tolerance threshold, the corresponding node to be evaluated is marked as a checkpoint.

[0032] And, generate an overall evaluation result based on the scores;

[0033] The report generation module is used to output an assessment report based on the overall assessment results generated by the quality assessment module.

[0034] Preferably, the anomaly detection module is used for:

[0035] The preset main time period is divided into multiple time segments. When the data fluctuation amplitude of a certain time segment exceeds a preset abnormal threshold, and the difference between the data fluctuation amplitude of that data segment and the data fluctuation amplitude of its adjacent time segment exceeds a preset difference threshold, an evaluation trigger event is identified.

[0036] Preferably, the quality assessment module is used for:

[0037] The data to be evaluated is scored based on the time distance between each node to be evaluated and the checkpoint, combined with preset weight adjustment rules.

[0038] Preferably, the quality assessment module is also used for:

[0039] Based on the time of the checkpoint, the offset time interval is defined by the set forward offset time and backward offset time.

[0040] Furthermore, identifying at least one node to be evaluated within the offset time interval includes:

[0041] Multiple nodes to be evaluated are determined based on preset node selection rules.

[0042] Preferably, the report generation module is used to generate and highlight a Level 1 warning message in the evaluation report when the overall evaluation result indicates that the composite statistical index is lower than the preset reference standard, so as to remind users or downstream systems to pay attention to data quality issues during that period. Beneficial effects

[0043] 1. This invention divides a preset main time period into time segments, identifies the data fluctuation amplitude within each time segment, and compares the data fluctuation amplitude of the current time segment with that of the adjacent time segments to obtain the difference. When the data fluctuation amplitude exceeds the abnormal threshold and the difference exceeds the difference threshold, the corresponding acquisition node is marked as a checkpoint. This allows for the screening of acquisition nodes caused by sudden changes in state from continuous geriatric syndrome data. It enables simultaneous attention to the absolute magnitude of data fluctuation amplitude and its rate of change over time, thereby effectively avoiding misjudgments caused by normal data fluctuations and improving the accuracy and relevance of the starting point and checkpoint positioning for quality assessment.

[0044] 2. This invention sets an offset time interval based on pre-defined checkpoints and identifies nodes to be evaluated within this offset time interval. Combining the time distance between each node to be evaluated and the checkpoint, it quantifies and scores the corresponding evaluation data according to preset weight adjustment rules. This extends the scope of quality evaluation from a single checkpoint to the offset interval before and after it, thereby enabling a centralized analysis of the contextual data of abnormal events. This invention also introduces a dynamic scoring mechanism linked to time distance, improving the accuracy of quality evaluation results by quantifying the correlation between different data collection nodes and core abnormal events within the interval.

[0045] 3. After scoring is completed, this invention identifies the node with the highest score as the reference node and uses its data as a comparison benchmark. Then, by performing difference calculation, when the difference between other data to be evaluated and the comparison benchmark exceeds the fluctuation tolerance threshold, the corresponding node to be evaluated is marked as a secondary checkpoint. This allows for the further discovery of potential, related secondary checkpoints within the focused offset time interval, using the node with the highest score as the reference node. This improves the ability to identify complex anomaly patterns and avoids evaluation bias caused by missed data collection nodes. Attached Figure Description

[0046] Figure 1 This is a flowchart of the method provided by the present invention;

[0047] Figure 2 This is a system module diagram provided by the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely for explaining the invention and are not intended to limit the scope of protection of the invention. Example

[0049] See Figure 1 This embodiment provides a quality assessment method based on geriatric syndrome data, including the following specific steps:

[0050] S1. Data acquisition and time segment division, the specific steps are as follows:

[0051] Before performing the quality assessment, acquire geriatric syndrome data collected within a preset subject time period; wherein: geriatric syndrome data includes a series of collection nodes arranged in chronological order and their corresponding collection data, such as continuous blood glucose monitoring data, heart rate variability data, or activity intensity data recorded by wearable devices;

[0052] The preset subject time period defines the complete time range of the analysis. For example, the complete time range is from midnight on the first day an elderly person begins to receive monitoring to the present time. This method ensures the closedness and completeness of the data analysis and enables refined analysis of the dynamic changes in the data.

[0053] The preset main time period is divided into multiple sequentially arranged time segments. Specifically, each time segment has an equal time length, such as 5 minutes or 10 minutes. By using an equal time length, a standardized data processing unit is provided for the subsequent establishment of preset anomaly judgment criteria, ensuring that the data fluctuation amplitude and data peak in different time segments are directly comparable.

[0054] Data is collected within each time segment to form the acquisition node, and each acquisition node records the physiological parameter value at a specific moment.

[0055] The preset main time period is set in advance to ensure the closedness and integrity of the data analysis. It is used to define the complete time range of data collection required for this quality assessment.

[0056] S2. Abnormal time node detection and calibration, the specific steps are as follows:

[0057] After dividing the time into segments, a preset numerical judgment criterion for identifying data mutations is used to detect abnormal time nodes in the data collection process, in order to identify sudden data changes that indicate potential health risks. The specific detection process is as follows:

[0058] Within each time segment, identify the data peak and the data fluctuation range. The data peak refers to the maximum value of the data collected within that time segment, while the data fluctuation range can be calculated as the absolute difference between the maximum and minimum values ​​of the data within that time segment, in order to quantify the degree of data dispersion within that time segment.

[0059] The fluctuation amplitude of the current time segment is compared with the fluctuation amplitude of one or more adjacent time segments to obtain the fluctuation amplitude difference, so as to distinguish between true instantaneous anomalies and continuous high fluctuation states.

[0060] Furthermore, when the data fluctuation amplitude of a time segment exceeds a preset abnormal threshold, and the difference in fluctuation amplitude exceeds a preset difference threshold, the time node corresponding to the data peak within this time segment is determined as an abnormal time node. This effectively filters out persistent high fluctuations and accurately identifies the moment when a sudden jump from a stable or low-fluctuation state to a high-fluctuation state. These moments are more strongly associated with acute health events such as falls and palpitations. Among them, persistent high fluctuations are caused by equipment interference or normal activities of the elderly, including turning over and walking.

[0061] Detected abnormal time points are marked as checkpoints, and pre-defined scoring criteria are associated with these checkpoints as the basis for subsequent in-depth analysis and scoring. The scoring criteria are specifically based on time distance and data characteristics for score calculation.

[0062] Furthermore, the numerical judgment criteria refer to a set of pre-defined logical rules and threshold combinations used to identify abnormal time nodes with sudden data changes from the data acquisition nodes.

[0063] S3. Determining the offset time interval and the node to be evaluated involves the following steps:

[0064] After calibrating the checkpoints, an offset time interval is determined based on the checkpoints to allow for a comprehensive evaluation of the context of the anomalous event. The offset time interval is used to define the time range of the data points to be evaluated. Specifically:

[0065] Based on the time of the checkpoint, the forward offset duration and the backward offset duration are set to jointly define the offset time interval. The forward offset duration and the backward offset duration can be set asymmetrically according to different physiological indicators or preset scenarios.

[0066] Furthermore, the important clinical significance of asymmetric settings is as follows: for a nocturnal hypoglycemic event, the trigger may have accumulated gradually over a long period of time before the event, such as 2-3 hours after dinner, so a longer forward offset duration is needed to trace it; while the blood glucose recovery period after the event may be shorter, so the backward offset duration can be set to be shorter.

[0067] Furthermore, within the determined offset time interval, multiple nodes to be evaluated are determined according to the preset node selection rules for sampling data points within the specified time interval.

[0068] The preferred node selection rule is to select nodes at fixed time intervals within the determined offset time interval.

[0069] The forward offset duration refers to the length of time that traces back from the checkpoint time, and is used to define the starting boundary of the offset time interval; the backward offset duration refers to the length of time that continues backward from the checkpoint time, and is used to define the ending boundary of the offset time interval.

[0070] S4. Data Scoring and Secondary Screening. After determining the nodes to be evaluated, the data evaluation and screening stage begins. The specific steps are as follows:

[0071] The collected data corresponding to the node to be evaluated is determined as the data to be evaluated. Based on the preset scoring criteria associated with the checkpoint in step S2, the data to be evaluated is scored to obtain the score of the data to be evaluated. The specific process includes:

[0072] Based on the time distance between each node to be evaluated and the checkpoint, an initial score is assigned to the data to be evaluated corresponding to that node using an initial score calculation method based on time distance. This initial score calculation method based on time distance is a calculation model used to calculate the initial score based on the time distance between the node to be evaluated and the checkpoint, and its specific definition is as follows:

[0073]

[0074] In the formula, This represents the initial score, which quantifies the initial importance of the data to be evaluated. This score decays exponentially as the time distance from the checkpoint increases.

[0075] This represents the maximum score, which is a system-preset constant representing the theoretically highest possible initial score.

[0076] This represents the time decay coefficient, a positive real number used to control the rate at which the score decays with increasing time distance. The larger the value, the faster the decay.

[0077] This represents the timestamp of the node to be evaluated, which means the absolute time when the node to be evaluated was collected;

[0078] This represents the checkpoint timestamp, which signifies the absolute time when a core anomaly event occurred.

[0079] According to the preset weight adjustment rules used to correct the initial score, the initial score is adjusted to obtain the final score of the data to be evaluated. The weight adjustment rules are used to reflect the correlation between the data points and the core abnormal events. Specifically, the weight adjustment rules are embodied in the calculation process. The weight adjustment coefficient output by the weight adjustment rules decreases monotonically with the increase of time distance. For example, if the calculation method of exponential decay or Gaussian decay is used, the node to be evaluated that is closer to the check point will get a higher weight adjustment coefficient and its final score will be higher.

[0080] The weight adjustment rule is a calculation model used to generate the weight adjustment coefficients, and its specific definition is as follows:

[0081]

[0082] In the formula, This represents the weight adjustment coefficient, which is a coefficient between 0 and 1 used to adjust the initial score. The closer a node is to the checkpoint, the closer its weight coefficient is to 1.

[0083] This represents the timestamp of the node to be evaluated, which means the absolute time when the node to be evaluated was collected;

[0084] This represents the checkpoint timestamp, which indicates the absolute time when the checkpoint occurred.

[0085] This represents the standard deviation, which is a parameter that controls the width of the weight decay. The larger the value, the slower the weight decays with distance, indicating that points over a larger time range are considered relevant.

[0086] After scoring is completed, the data to be evaluated are sorted in descending order according to their scores, and the node corresponding to the data with the highest score is selected and determined as the reference node.

[0087] The data to be evaluated corresponding to the reference node is determined as the comparison benchmark. The difference between the other data to be evaluated and the comparison benchmark is calculated to obtain the data difference. It is then determined whether the data difference exceeds the preset fluctuation tolerance threshold. The fluctuation tolerance threshold represents the normal fluctuation range that data points should have within the same abnormal event's influence range.

[0088] When the data difference exceeds the preset fluctuation tolerance threshold, the data point exhibits new independent abnormal characteristics relative to the core abnormal point. The data to be evaluated corresponding to the data difference is judged as abnormal data, and the corresponding node to be evaluated is marked as a secondary checkpoint. Based on the secondary checkpoint, the process returns to determine the offset time interval and subsequent scoring and screening steps to perform recursive in-depth analysis of complex or chain-like abnormal events.

[0089] When the data difference does not exceed the preset fluctuation tolerance threshold, the data to be evaluated corresponding to the data difference is judged as normal data. The offset direction of the normal data relative to the comparison benchmark is recorded, that is, whether it is greater than or less than the comparison benchmark. This information is determined as the new offset analysis starting point. This information is recorded and used in the final evaluation report to generate a quantitative description of the direction of health recovery or the trend of deterioration.

[0090] S5. Generate overall assessment results and report. After scoring one or more checkpoints and all associated data to be assessed, generate overall assessment results. The specific steps are as follows:

[0091] Calculate the average score of all data to be evaluated, and count the total number of nodes to be evaluated within the offset time interval;

[0092] The average score and the total number of nodes to be evaluated are combined to form a composite statistical index to more comprehensively characterize the data quality. For example, a high average score accompanied by a large total number of nodes to be evaluated indicates that the data quality is stable and information-rich before and after the abnormal event; while a high average score but a small total number of nodes to be evaluated may indicate insufficient data collection or that most data was judged as abnormal data due to secondary screening.

[0093] The composite statistical indicators are compared with the preset reference standards to generate an overall evaluation result. The reference standards refer to the preset benchmark values ​​or benchmark ranges used to compare with the composite statistical indicators to determine the data quality level.

[0094] Based on the scores and time sequence of the data to be evaluated, an abnormal change trend is generated, and an evaluation report containing the overall evaluation results and the abnormal change trend is output. The abnormal change trend is a structured description generated based on the scores and time sequence of the data to be evaluated, which is used to visualize or quantify the dynamic evolution of the data after the occurrence of an abnormal event. Preferably, it can be a score sequence graph sorted by time, or a quantitative indicator describing the time required for the score to fall from the peak to a stable state.

[0095] Furthermore, when the composite statistical indicator is lower than the preset reference standard, a level one early warning message is generated in the evaluation report, along with a preset explanation of the reason for the early warning.

[0096] The warning was explained as follows: after an abnormal event occurred, data recovery was slow and fluctuated frequently.

[0097] The Level 1 warning information refers to the explicit risk warning information generated in the assessment report when the composite statistical indicator is lower than the reference standard. The warning reason explanation refers to the pre-set text associated with the Level 1 warning information, which explains the specific reasons that triggered the warning. Example

[0098] See Figure 2 This embodiment provides a quality assessment system based on geriatric syndrome data, including the following modules:

[0099] The anomaly detection module is configured to acquire geriatric syndrome data collected within a preset main time period, continuously monitor the data to identify collection nodes that meet preset anomaly judgment criteria, and generate assessment trigger events.

[0100] Furthermore, the preset main time period is divided into multiple consecutive time segments according to a fixed time length. For each time segment, the data fluctuation range of its internal data is calculated, such as calculating the standard deviation or range of the data values ​​of all collection nodes within that time segment.

[0101] Furthermore, the data fluctuation amplitude of the current time segment is compared with a preset anomaly threshold; the difference between the data fluctuation amplitude of the current time segment and the fluctuation amplitude of its adjacent previous or next time segment is calculated.

[0102] The difference is compared with a preset difference threshold: when the data fluctuation of a certain time segment exceeds the preset anomaly threshold and the difference in fluctuation exceeds the preset difference threshold, it is determined that there is a data acquisition node in the time segment that meets the preset anomaly judgment criteria, an evaluation trigger event is generated, and the evaluation trigger event and related data acquisition node information are transmitted to the quality evaluation module.

[0103] The quality assessment module is used to respond to assessment trigger events generated by the anomaly detection module and execute the core quality assessment process, which specifically includes:

[0104] Upon receiving an evaluation trigger event, the data collection node specified in the event that meets the preset anomaly judgment criteria is marked as a checkpoint;

[0105] Based on the timestamp of the checkpoint, the offset time interval is defined by the set forward offset duration and backward offset duration. Specifically, if the forward offset duration is 30 minutes and the backward offset duration is 30 minutes, then the offset time interval is the range of 30 minutes before and after the checkpoint.

[0106] Within this offset time interval, at least one node to be evaluated is determined according to a preset node selection rule. The node selection rule can be uniform sampling within the interval or selecting all sampling nodes.

[0107] The collected data corresponding to these nodes to be evaluated is determined as the data to be evaluated. Each piece of data to be evaluated is scored according to the preset scoring criteria. Specifically, the time distance between each node to be evaluated and the checkpoint is calculated. Based on the time distance, the corresponding data to be evaluated is assigned an initial score. For example, the closer the time distance, the higher the initial score.

[0108] The initial score is adjusted by applying a preset weight adjustment rule to obtain the final score. This weight adjustment rule can take into account the numerical value or other statistical characteristics of the data to be evaluated, so as to make the scoring more comprehensive.

[0109] After scoring all the data to be evaluated, an internal verification process is executed. The node corresponding to the highest final score among all the data to be evaluated is determined as the reference node, and the data value of the reference node is used as the comparison benchmark. For all other data to be evaluated, the difference between them and the comparison benchmark is calculated. If the calculated data difference exceeds the preset fluctuation tolerance threshold, it indicates that the data point is too different from the most reliable data point in the interval. The node to be evaluated corresponding to this data difference is also marked as a secondary checkpoint to trigger a new round of more local quality assessment.

[0110] Based on the final scores of all the data to be evaluated, their average scores are calculated, and combined with the total number of nodes to be evaluated, a composite statistical index is generated. The composite statistical index is used to compare with a preset reference standard to determine the overall quality level of the data within the offset time interval. Finally, an overall evaluation result containing the composite statistical index and the comparison results is generated and output.

[0111] The report generation module is configured to receive the overall assessment results generated by the quality assessment module and output a structured assessment report based on the overall assessment results. The assessment report clearly lists the checkpoints that triggered the assessment, the offset time intervals covered by the assessment, the final scores of each data to be assessed, and the final generated composite statistical indicators, as well as other information related to this quality assessment.

[0112] When the overall assessment results indicate that the composite statistical indicators are lower than the preset reference standards, a Level 1 warning message will be generated and highlighted in the assessment report to remind users or downstream systems to pay attention to data quality issues during that period and to suggest data cleaning, correction, or manual review.

Claims

1. A quality assessment method based on geriatric syndrome data, characterized in that, When a data collection node that meets the preset abnormality criteria is detected in the geriatric syndrome data collected within the preset main time period, a quality assessment is performed. The quality assessment includes: Based on the acquisition nodes that meet the preset anomaly judgment criteria, they are marked as checkpoints and the offset time interval is determined; Within the offset time interval, at least one node to be evaluated is determined, and the collected data corresponding to the node to be evaluated is determined as the data to be evaluated. And score the data to be evaluated according to the preset scoring criteria to generate an overall evaluation result, and output an evaluation report based on the overall evaluation result; Among them, those that meet the preset anomaly judgment criteria include: After dividing the preset main time period into multiple time segments, if the data fluctuation amplitude of a certain time segment exceeds the preset abnormal threshold, and the difference between the data fluctuation amplitude of that time segment and the data fluctuation amplitude of its adjacent time segments exceeds the preset difference threshold; Determining at least one node to be evaluated within the offset time interval includes: Multiple nodes to be evaluated are determined according to preset node selection rules; The scoring of the data to be evaluated, based on the preset scoring criteria, includes: An initial score is assigned to the data to be evaluated corresponding to each node based on the time distance between each node to be evaluated and the checkpoint. In addition, the initial scores are adjusted according to the preset weight adjustment rules to obtain the scores of the data to be evaluated.

2. The quality assessment method based on geriatric syndrome data according to claim 1, characterized in that, The process of identifying checkpoints and determining the offset time interval includes: Based on the time of the checkpoint, the offset time interval is defined by the set forward offset time and backward offset time.

3. The quality assessment method based on geriatric syndrome data according to claim 1, characterized in that, The method further includes: After scoring the data to be evaluated, the node corresponding to the data with the highest score is determined as the reference node, and its data is determined as the benchmark. The other data to be evaluated are compared with the benchmark, and when the difference exceeds the preset fluctuation tolerance threshold, the node corresponding to the difference is marked as a checkpoint to trigger the quality assessment.

4. The quality assessment method based on geriatric syndrome data according to claim 1, characterized in that, The evaluation process involves scoring the data to be evaluated according to a preset scoring standard to generate an overall evaluation result. This includes generating statistical indicators based on the average score of all data to be evaluated and the total number of nodes to be evaluated; comparing the statistical indicators with preset reference standards; and generating a level-one warning message in the evaluation report when the statistical indicators are lower than the reference standards.

5. A quality assessment system based on geriatric syndrome data, characterized in that, include: The anomaly detection module is used to acquire geriatric syndrome data collected within a preset main time period and monitor the data to identify collection nodes that meet preset anomaly judgment criteria, which are then used as evaluation trigger events. The quality assessment module is used to perform quality assessments for assessment trigger events identified by the anomaly detection module. The quality assessment module is configured as follows: The acquisition nodes that meet the preset anomaly judgment criteria are marked as checkpoints, and the offset time interval is determined based on the checkpoints; Within the offset time interval, the node to be evaluated is determined, and the data to be evaluated corresponding to the node to be evaluated is scored according to the preset scoring criteria. After scoring the data to be evaluated, the node corresponding to the data with the highest score is determined as the reference node, and the other data to be evaluated are compared with the benchmark corresponding to the reference node. When the comparison result exceeds the preset fluctuation tolerance threshold, the corresponding node to be evaluated is marked as a checkpoint. And, generate an overall evaluation result based on the scores; The report generation module is used to output an assessment report based on the overall assessment results generated by the quality assessment module. The anomaly detection module is used to: divide the preset main time period into multiple time segments, and when the data fluctuation amplitude of a certain time segment exceeds the preset anomaly threshold, and the difference between the data fluctuation amplitude of the data segment and the data fluctuation amplitude of its adjacent time segment exceeds the preset difference threshold, identify the evaluation trigger event; The quality assessment module is used to score the data to be assessed based on the time distance between each node to be assessed and the checkpoint, and in conjunction with preset weight adjustment rules.

6. A quality assessment system based on geriatric syndrome data according to claim 5, characterized in that, The quality assessment module is also used for: Based on the time of the checkpoint, the offset time interval is defined by the set forward offset time and backward offset time. Furthermore, identifying at least one node to be evaluated within the offset time interval includes: Multiple nodes to be evaluated are determined based on preset node selection rules.

Images