Cerebral stroke cognitive impairment risk auxiliary assessment method and cerebral stroke cognitive impairment risk auxiliary assessment system

Abstract

The invention relates to the technical field of medical information processing, in particular to a cerebral apoplexy cognitive impairment risk auxiliary assessment method and a cerebral apoplexy cognitive impairment risk auxiliary assessment system. According to the method, various monitoring time sequence data and the average diffusivity in the brain nuclear magnetic resonance image result are obtained, and dynamic changes in the data and differences among different individuals can be better adapted through adaptive segmented monitoring of the time sequence data. And determining an influence coefficient by combining the similarity between the data segments and the average diffusivity. And calculating cognitive ability decline and intervention indexes based on the change trend and influence coefficient of the data segment, and dynamically reflecting the cognitive change of the patient. The evaluation mode provides accurate data for doctors, and assists in real-time monitoring of cognitive decline and timely intervention.

Description

Technical Field

[0001] The present invention relates to the technical field of medical information processing, and in particular to a method and system for auxiliary assessment of the risk of cognitive impairment in stroke. Background Art

[0002] Stroke is a common cerebrovascular disease with a high incidence rate and often has a serious impact on the patient's cognitive function. Cognitive impairment is a common complication after stroke, which has a significant impact on the patient's daily living ability and quality of life. Therefore, auxiliary assessment of the risk of cognitive impairment in stroke patients is of great significance for doctors to formulate personalized treatment plans and rehabilitation plans for patients. In the context of current medical technology, the application of artificial intelligence technology has brought new possibilities in this field.

[0003] Existing technologies usually use artificial intelligence technology to assist in the assessment of the risk of cognitive impairment in stroke patients, such as using an assessment model to assist in the assessment of cognitive impairment based on the patient's monitoring data. However, the patient's cognitive ability after a stroke will have a trend of periodic decline or recovery, so cognitive impairment in stroke patients is a complex and dynamically changing process, and there will be individual differences. Therefore, using a fixed assessment model to assist in the assessment of the risk of cognitive impairment in stroke patients will make the auxiliary assessment results inaccurate. Summary of the invention

[0004] In order to solve the technical problem that cognitive impairment in stroke patients is a complex and dynamically changing process and there are individual differences, the use of a fixed assessment model to assist in assessing the risk of cognitive impairment in stroke patients will make the auxiliary assessment results inaccurate, the purpose of the present invention is to provide a method and system for assisting in assessing the risk of cognitive impairment in stroke. The technical solutions adopted are as follows:

[0005] Obtaining various monitoring time series data of stroke subjects and average diffusion rate in brain magnetic resonance imaging results;

[0006] Based on the change of data values ​​in each monitoring time series data, each monitoring time series data is adaptively segmented to obtain multiple monitoring data segments in each monitoring time series data; analyzing the similarity of changes between data values ​​in monitoring data segments of different monitoring time series data, and determining the influence coefficient of each monitoring time series data in combination with the average diffusion rate;

[0007] In all monitoring time series data, based on the data change trend of the monitoring data segment at each moment and the influence coefficient of each monitoring time series data, the cognitive ability decline index of the subject to be tested at each moment is determined; based on the difference between the cognitive ability decline index of the subject to be tested at different moments and the interval between moments, the intervention index of the subject to be tested that needs to be intervened at each moment is calculated;

[0008] Based on the cognitive ability decline index of the subject at all times and the intervention index requiring intervention, an auxiliary assessment of the cognitive impairment risk of the subject is performed.

[0009] Furthermore, the method for acquiring the monitoring data segment includes:

[0010] Each monitoring time series data is segmented based on an adaptive piecewise constant approximation algorithm to obtain an initial data segment;

[0011] In each monitoring time series data, the variation range of the data value in each initial data segment is analyzed to obtain the variation characteristic value of each initial data segment;

[0012] Obtain the curve corresponding to each initial data segment, and take the average of the tangent slope values ​​at all data values ​​in each initial data segment as the trend characteristic value of each initial data segment;

[0013] For any two adjacent initial data segments in each monitoring time series data, the absolute value of the difference between the change characteristic values ​​is used as the change difference factor, and the absolute value of the sum of the trend characteristic values ​​is used as the trend similarity factor. According to the change difference factor and the trend similarity factor, the merging index between the two adjacent initial data segments is calculated, and the change difference factor is negatively correlated with the merging index, and the trend similarity factor is positively correlated with the merging index;

[0014] In each monitoring time series data, the first initial data segment is taken as the data segment to be tested to start traversal, and it is determined whether the merging index between the data segment to be tested and the next adjacent initial data segment meets the preset condition. If not, the current data segment to be tested is taken as a monitoring data segment, and the next adjacent initial data segment is taken as a new data segment to be tested to continue traversal;

[0015] If the condition is satisfied, the next adjacent initial data segment is merged into the data segment to be tested to obtain a new data segment to be tested, and it is determined whether the merging index between the last initial data segment in the new data segment to be tested and the next adjacent initial data segment meets the preset condition. If not, the current data segment to be tested is used as a detection data segment, and the next adjacent initial data segment is used as a new data segment to be tested to continue traversing;

[0016] After the traversal is completed, all monitoring data segments of each monitoring time series data are obtained;

[0017] The preset condition is that the merging index is greater than or equal to a preset merging threshold.

[0018] Furthermore, the method for obtaining the change characteristic value includes:

[0019] In each initial data segment, the difference between the maximum data value and the mean of all data values ​​is taken as the first fluctuation factor of each initial data segment, the variance of all data values ​​is taken as the second fluctuation factor, and the sum of the first fluctuation factor and the second fluctuation factor is taken as the change characteristic value of each initial data segment.

[0020] Furthermore, the method for obtaining the influence coefficient includes:

[0021] In each monitoring time series data, based on the difference between the change characteristic value of the initial data segment in each monitoring data segment and the change characteristic values ​​of all initial data segments in each monitoring time series data, determine the change degree value of each monitoring data segment;

[0022] Determine the comparison data segment of each monitoring data segment in each monitoring time series data in other monitoring time series data;

[0023] In each monitoring time series data, the DTW values ​​between each monitoring data segment and all corresponding comparison data segments are comprehensively analyzed to determine the first impact factor of each monitoring data segment; the difference between the change degree values ​​of each monitoring data segment and all corresponding comparison data segments is comprehensively analyzed to obtain the second impact factor of each monitoring data segment;

[0024] According to the first influencing factor, the second influencing factor and the average diffusion rate, an influencing characteristic value of each monitoring data segment is obtained, wherein the first influencing factor and the second influencing factor are both negatively correlated with the influencing characteristic value, and the average diffusion rate is positively correlated with the influencing characteristic value;

[0025] In each monitoring time series data, the sum of the influencing characteristic values ​​of all monitoring data segments is normalized and used as the influencing coefficient of each monitoring time series data.

[0026] Furthermore, the method for obtaining the comparison data segment includes:

[0027] In each monitoring data segment of each monitoring time series data, the mean value of all data values ​​at corresponding moments is used as the reference value of each monitoring data segment;

[0028] In any type of monitoring time series data, any monitoring data segment is selected as the target data segment. In each other monitoring time series data, the monitoring data segment with the smallest absolute difference with the reference value of the target data segment is selected as the comparison data segment of the target data segment in each other monitoring time series data.

[0029] Furthermore, the types of monitoring time series data include at least glycosylated hemoglobin level, MoCA scale score and MMSE scale score.

[0030] Furthermore, the method for obtaining the cognitive ability decline index includes:

[0031] In each monitoring time series data, the cognitive ability decline factor of each monitoring data segment is determined based on the data change trend of the monitoring data segment;

[0032] In each monitoring time series data, the cognitive ability decline characteristic value is determined according to the difference between the cognitive ability decline factor of each monitoring data segment and the cognitive ability decline factor of the first monitoring data segment;

[0033] The product of the cognitive ability decline characteristic value of each monitoring data segment and the corresponding influence coefficient is taken as the cognitive ability decline coefficient of each monitoring data segment; the sum of the cognitive ability decline coefficients of the monitoring data segment in all types of monitoring time series data at each moment is normalized and used as the cognitive ability decline index of the subject to be tested at each moment.

[0034] Furthermore, the method for obtaining the cognitive ability decline factor includes:

[0035] Among all the initial data segments, the initial data segments with positive trend characteristic values ​​are regarded as rising data segments, and the initial data segments with negative trend characteristic values ​​are regarded as falling data segments;

[0036] In each rising data segment, the difference between the data values ​​of the right endpoint and the left endpoint is used as the rising factor, and the sum of the rising factors of all rising data segments is used as the rising characteristic value;

[0037] In each descending data segment, the difference between the data values ​​of the left endpoint and the right endpoint is used as the descending factor, and the sum of the descending factors of all descending data segments is used as the descending characteristic value;

[0038] In each monitoring data segment, the ratio between the rising characteristic value and the falling characteristic value is used as the rising relative value of each monitoring data segment; the ratio between the number of rising data segments and the total number of initial data segments in each monitoring data segment is used as the quantity factor; the product of the rising relative value and the quantity factor is used as the rising coefficient of each monitoring data segment;

[0039] Determine the relationship between each monitoring time series data and cognitive decline. If it is a positive correlation, the normalized value of the rising coefficient of the monitoring data segment is used as the cognitive decline factor. If it is a negative correlation, the normalized value of the inverse of the rising coefficient of the monitoring data segment is used as the cognitive decline factor.

[0040] Furthermore, the method for obtaining the intervention index includes:

[0041] The time interval between each moment and the first moment is taken as the moment difference value; the normalized value of the difference between the cognitive ability decline index at the first moment and each moment is taken as the cognitive recovery index;

[0042] According to the cognitive ability decline index, moment difference value and cognitive recovery index corresponding to each moment, the intervention index required for the subject to be tested at each moment is determined, and the cognitive ability decline index is positively correlated with the intervention index, and the moment difference value and cognitive recovery index are negatively correlated with the intervention index.

[0043] The present invention also proposes a system for auxiliary assessment of the risk of cognitive impairment in stroke, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any step of the method for auxiliary assessment of the risk of cognitive impairment in stroke when executing the computer program.

[0044] The present invention has the following beneficial effects:

[0045] By obtaining a variety of monitoring time series data of the stroke object to be tested and the average diffusion rate in the brain magnetic resonance imaging results, and combining these data for analysis, the cognitive function status of the object to be tested can be more comprehensively reflected, thereby improving the accuracy of the auxiliary assessment of cognitive impairment risk. In the present invention, adaptive segmentation is performed based on the change of data values ​​in each monitoring time series data to obtain monitoring data segments; at the same time, the similarities between the monitoring data segments in different monitoring time series data are analyzed, and the influence coefficient of each monitoring time series data is determined in combination with the average diffusion rate. This step makes the evaluation process more personalized and can better adapt to the dynamic changes in the data and the differences between different individuals. Based on the data change trend and influence coefficient of the monitoring data segment, the cognitive ability decline index of the object to be tested at each moment is determined, and the intervention index that needs to be intervened at each moment is calculated based on the cognitive ability decline index. This dynamic evaluation method can more accurately reflect the development and change of the patient's cognitive ability, provide doctors with more accurate auxiliary evaluation data, and enable doctors to dynamically monitor the cognitive decline of the object to be tested in the subsequent process, timely discover and take intervention measures, so as to formulate more personalized treatment plans and rehabilitation plans for the object to be tested, and improve the treatment effect and quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A method flow chart of a method for assisting in assessing the risk of cognitive impairment in stroke provided by one embodiment of the present invention;

[0048] Figure 2 A method flow chart of a method for acquiring monitoring data segments provided by one embodiment of the present invention;

[0049] Figure 3 A method flow chart of a method for obtaining an influence coefficient provided by an embodiment of the present invention;

[0050] Figure 4 A method flow chart of a method for obtaining a cognitive ability decline index provided by one embodiment of the present invention;

[0051] Figure 5 A system block diagram of a stroke cognitive impairment risk auxiliary assessment system provided by an embodiment of the present invention;

[0052] Figure 6 A schematic diagram of the system structure of a stroke cognitive impairment risk auxiliary assessment system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method and system for auxiliary assessment of cognitive impairment risk in stroke proposed according to the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0054] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0055] A specific solution provided by the present invention is described in detail below with reference to the accompanying drawings.

[0056] See also Figure 1, which shows a method flow chart of a method for auxiliary assessment of cognitive impairment risk of stroke provided by an embodiment of the present invention, the method comprising the following steps:

[0057] Step S1: obtaining a variety of monitoring time series data of the stroke subject and the average diffusion rate in the brain magnetic resonance imaging results.

[0058] Post-stroke cognitive impairment (PSCI) is a clinical syndrome that occurs after a stroke and is characterized by cognitive impairment in memory, thinking, etc. PSCI diagnosis is often determined by cognitive assessment after a stroke. It is triggered by a stroke event, and cognitive changes after triggering are a dynamic process. Its occurrence and development are somewhat hidden. If it is not discovered and intervened in time, it may lead to aggravated cognitive decline, resulting in consequences such as post-stroke dementia and increased disability rate. In order to more promptly discover and intervene in stroke subjects in a timely manner, advanced artificial intelligence technology is usually used to enable the assessment method to better adapt to the dynamic changes in patients' cognitive abilities, which helps to more accurately capture the risk factors of patients' cognitive impairment, thereby providing doctors with more scientific and personalized treatment recommendations and rehabilitation programs.

[0059] In an embodiment of the present invention, by acquiring a variety of monitoring time series data of the stroke subject and the average diffusion rate in the brain magnetic resonance imaging results, the dynamic changes in the cognitive ability of the subject are monitored, thereby providing doctors with more accurate auxiliary evaluation data when making a diagnosis or formulating a plan.

[0060] In view of the fact that high blood sugar state can affect the normal function of the nervous system and lead to impaired cognitive function, and the glycated hemoglobin level is an important indicator for evaluating blood sugar control, in the embodiment of the present invention, the glycated hemoglobin level is used as a monitoring data, and the glycated hemoglobin level of the stroke subject is obtained by a glycated hemoglobin analyzer to evaluate the cognitive decline of the stroke patient.

[0061] At the same time, since the Montreal Cognitive Assessment (MoCA) includes 8 cognitive areas of visual space and executive function, naming, memory, attention, language, abstraction, delayed recall and orientation, it can comprehensively evaluate the patient's cognitive function, and has high sensitivity and specificity for identifying mild cognitive impairment (MCI) and dementia, and is particularly suitable for screening and evaluation of post-stroke cognitive impairment; and the Mini-Mental State Examination (MMSE) is sensitive to memory and language (left hemisphere stroke), and has high sensitivity and specificity for dementia diagnosis. Therefore, in this embodiment of the present invention, the MoCA scale score and the MMSE scale score are also used as a monitoring data to evaluate the cognitive decline of the stroke patient to be tested.

[0062] Because the temporal relationship between stroke and cognitive impairment is that it occurs after the stroke event and lasts for 3 to 6 months, the monitoring time series data of various monitoring data are obtained in weeks. The average diffusion rate in the magnetic resonance imaging results is used to reflect the degree of brain damage. Since the degree of brain damage will not vary significantly in a short period of time, it only needs to be obtained once, that is, the average diffusion rate in the brain magnetic resonance imaging results of the stroke subject on the day of the stroke.

[0063] It should be noted that the monitoring interval of the monitoring data can also be in units of months, 3 days, or 5 days, etc., which can be adjusted according to the implementation scenario and is not limited here; and the length of the time series data can also vary according to the implementation scenario and is not limited here. In other embodiments of the present invention, other types of monitoring data can also be used, such as: blood lipid levels, inflammatory markers, homocysteine, NIHSS scores, etc., and the specific types are not limited here.

[0064] In the embodiment of the present invention, the collection and acquisition of various data of the stroke test subjects are all authorized by the relevant users, and the process does not violate relevant laws and regulations and does not violate public order and good customs.

[0065] Step S2: Based on the change of data values ​​in each monitoring time series data, each monitoring time series data is adaptively segmented to obtain multiple monitoring data segments in each monitoring time series data; the similarity of changes between data values ​​in monitoring data segments of different monitoring time series data is analyzed, and combined with the average diffusion rate, the influence coefficient of each monitoring time series data is determined.

[0066] Because the cognitive decline of the subjects to be tested is in a dynamic process, the data values ​​in the monitoring time series data will also change dynamically accordingly. Therefore, by adaptively segmenting the data values ​​based on the changes in each monitoring time series data, it is helpful to decompose the original continuous and huge data stream into smaller units that are easier to analyze and process - monitoring data segments, so as to more finely capture the fluctuations and change patterns in the data in each monitoring data segment. At the same time, since there are certain correlations and influences between different types of monitoring data, they jointly reflect the cognitive status of the subjects to be tested, so we continue to analyze the similarity of changes between data values ​​in monitoring data segments in different detection time series data, and combine the average diffusion rate in the brain magnetic resonance imaging results to quantify the influence coefficient of each monitoring time series data, which is used to reflect the potential correlation or potential influence of each monitoring data on other monitoring data, providing an important basis for the subsequent evaluation process.

[0067] Cognitive impairment after stroke is triggered by a stroke event, and the process of cognitive decline is a dynamic process. Specifically, after a stroke, cognitive ability may decline rapidly in the short term. As the disease recovers, cognitive ability will gradually improve. In the absence of cognitive impairment, cognitive level will gradually recover and approach normal level. However, when cognitive impairment occurs, cognitive level will decline again after a short recovery, and cognitive level will seriously decline. In addition, various human indicators fluctuate over time. For example, the glycated hemoglobin levels obtained at different times may vary. The MoCA score index and the MMSE score index are psychological scale test results. Their values ​​vary according to the current state, degree of cooperation and test environment of the subject. Therefore, in order to more accurately obtain the change results of each monitoring data at different stages, based on the change of data values ​​in each monitoring time series data, it is adaptively segmented to obtain multiple monitoring data segments in each monitoring time series data.

[0068] Preferably, in one embodiment of the present invention, the method for acquiring the monitoring data segment includes:

[0069] See also Figure 2 , which shows a method flow chart of a method for acquiring monitoring data segments in one embodiment of the present invention, the method comprising the following steps:

[0070] Step S201: Initially divide each type of monitoring time series data to obtain initial data segments.

[0071] Adaptive Piecewise Constant Approximation (APCA) is a time series data analysis method that adaptively segments time series data according to the changes in the data, and makes the data changes in each data segment relatively consistent, so as to better capture the local characteristics of the data. Therefore, each monitoring time series data is segmented based on the adaptive piecewise constant approximation algorithm to obtain initial data segments; the initial data segments obtained by this method can initially and automatically adapt to the actual changes in the data, avoiding the inaccuracy of segmentation by the preset method.

[0072] It should be noted that the APCA algorithm is a technical means well known to technical personnel in this field and will not be described in detail here. The initial strategy is random initialization, the number of iterations is set to 100 times, and the specific settings of various parameters in the algorithm can be adjusted according to the implementation scenario, which will not be described or limited here.

[0073] Step S202: In each monitoring time series data, the variation range and variation trend of the data value in each initial data segment are analyzed to obtain the variation characteristic value and trend characteristic value of each initial data segment.

[0074] Since the maximum value in a set of data represents an extreme level of the data, in each initial data segment, the difference between the maximum data value and the mean of all data values ​​is taken as the first fluctuation factor of each initial data segment, and the variance of all data values ​​is taken as the second fluctuation factor; the larger the first fluctuation factor is, the larger the fluctuation range of the data values ​​in the initial data segment is, and the larger the second fluctuation factor is, the more drastic the fluctuation of the data values ​​in the initial data segment is, so the sum of the first fluctuation factor and the second fluctuation factor is taken as the change characteristic value of each initial data segment, denoted as HT; the larger the change characteristic value is, the more drastic the fluctuation of the data values ​​in the initial data segment is, and at the same time, the larger the fluctuation range is.

[0075] In other embodiments of the present invention, the difference between the maximum value and the minimum value in each initial data segment can also be used as the first fluctuation factor of each initial data segment. Similarly, the larger the first fluctuation factor is, the greater the fluctuation range of the data value in the initial data segment is.

[0076] Then obtain the curve corresponding to each initial data segment (smoothly connect each data value in time series), and calculate the tangent slope value at each data value on the curve, and take the average of the tangent slope values ​​at all data values ​​in each initial data segment as the trend characteristic value of each initial data segment, denoted as QT; the trend characteristic value characterizes the changing trend of the data values ​​in each initial data segment. If the value is positive and the larger it is, the overall changing trend is increasing. Conversely, if the value is negative and the smaller it is, the overall changing trend is decreasing.

[0077] Step S203: In each monitoring time series data, based on the change characteristic values ​​and trend characteristic values ​​of adjacent initial data segments, a merging index is obtained, and adjacent initial data segments are merged based on the merging index to obtain all monitoring data segments in each monitoring time series data.

[0078] Since both the change characteristic value and the trend characteristic value represent the change characteristics of the data in the initial data segment, the similarity between these two indicators of adjacent initial data segments can be analyzed, so as to merge the initial data segments and obtain the monitoring data segments.

[0079] For any two adjacent initial data segments in each monitoring time series data, first calculate the difference between the change feature values, and use the absolute value of the difference as the change difference factor; then use the absolute value of the sum of the trend feature values ​​as the trend similarity factor. Both the change difference factor and the trend similarity factor can characterize the similarity of the data change characteristics between two adjacent initial data segments. Then, combine the change difference factor and the trend similarity factor to calculate the merger index between the two adjacent initial data segments. The formula model of the merger index can be, for example, as follows:

[0080]

[0081] Among them, HB represents the merging index between two adjacent initial data segments; HT1 represents the change characteristic value of initial data segment 1 among two adjacent initial data segments; HT2 represents the change characteristic value of initial data segment 2 among two adjacent initial data segments; QT1 represents the trend characteristic value of initial data segment 1 among two adjacent initial data segments; QT2 represents the trend characteristic value of initial data segment 2 among two adjacent initial data segments; norm() represents the normalization function; γ1 represents the preset first parameter.

[0082] In the formula model of the merger index, when the change difference factor |HT1-HT2| between two adjacent initial data segments is larger, it means that the two have a less consistent change range, so the merger index should be reduced; since the trend characteristic value may be positive or negative, if the trend similarity factor |QT1+QT2| between two adjacent initial data segments is larger, it means that the two have a more consistent change trend, so the merger index should be increased. Therefore, the change difference factor is placed in the denominator and the trend similarity factor is placed in the numerator, so that the ratio is normalized to obtain the merger index. At this time, the larger the merger index, the more it is necessary to merge the two adjacent initial data segments.

[0083] It should be noted that the purpose of presetting the first parameter γ1 is to prevent the denominator from being 0, and the value may be 0.001. The specific value may be adjusted according to the implementation scenario and is not limited here.

[0084] After obtaining the merging index, the initial data segments can be merged based on this index to obtain monitoring data segments. In each monitoring time series data, the first initial data segment is used as the data segment to be tested to start traversal, and it is determined whether the merging index between the data segment to be tested and the next adjacent initial data segment meets the preset conditions. If not, the current data segment to be tested is used as a monitoring data segment, and the next adjacent initial data segment is used as a new data segment to be tested to continue traversal; if it is satisfied, the next adjacent initial data segment is merged into the data segment to be tested to obtain a new data segment to be tested, and it is determined whether the merging index between the last initial data segment in the new data segment to be tested and the next adjacent initial data segment meets the preset conditions. If not, the current data segment to be tested is used as a monitoring data segment, and the next adjacent initial data segment is used as a new data segment to be tested to continue traversal; after the traversal is completed, all monitoring data segments in each monitoring time series data are obtained.

[0085] The preset condition is that the merging index is greater than or equal to a preset merging threshold.

[0086] It should be noted that the preset merging threshold is set to 0.6, and the specific value can be adjusted according to the implementation scenario and is not limited here.

[0087] An example is given to illustrate the acquisition process of the above monitoring data segments: assuming that there are 4 initial data segments in a certain monitoring time series data, which are recorded as initial data segment 1, initial data segment 2, initial data segment 3, and initial data segment 4 respectively. For initial data segment 1, determine whether the merging index between it and initial data segment 2 meets the preset condition. If so, initial data segment 1 and initial data segment 2 form a new data segment a to be tested; then determine whether the merging index between initial data segment 2 and initial data segment 3 in the new data segment a to be tested meets the preset condition. If not, the new data segment a to be tested composed of initial data segment 1 and initial data segment 2 is used as a monitoring data segment, and initial data segment 3 is used as a new data segment to be tested. Continue to determine whether the merging index between initial data segment 3 and initial data segment 4 meets the preset condition. If not, initial data segment 3 is used as a monitoring data segment, and initial data segment 4 is used as a new data segment to be tested. At this time, since initial data segment 4 is the last initial data segment in the monitoring time series data, it becomes a monitoring data segment alone. At this point, the four initial data segments are merged to obtain three monitoring data segments.

[0088] During the cognitive decline or recovery of stroke patients, monitoring data will fluctuate dynamically, and different monitoring data will often have a tendency to influence each other or change together, which is specifically manifested as follows: increased levels of glycosylated protein will increase the risk of cerebrovascular injury; stroke events will cause cerebrovascular damage, focal or overall brain tissue damage, and increase the average diffusion rate of magnetic resonance diffusion tensor imaging compared with normal levels, increasing the possibility of cognitive impairment in patients; MoCA scale and MMSE scale tests reflect the current subjective cognitive ability of the subject, and the lower the score, the worse the patient's cognitive ability. Therefore, by analyzing the similarity of changes in data values ​​of monitoring data segments in different monitoring time series data, and combining the average diffusion rate in brain magnetic resonance imaging results, the influence coefficient of each monitoring time series data can be determined, thereby revealing the potential connection between different monitoring time series data and providing data support for subsequent analysis results.

[0089] Preferably, in one embodiment of the present invention, the method for obtaining the influence coefficient includes:

[0090] See also Figure 3 , which shows a method flow chart of a method for obtaining an influence coefficient in one embodiment of the present invention, the method comprising the following steps:

[0091] Step S211: in each monitoring time series data, determine the change degree value of each monitoring data segment based on the difference between the change characteristic values ​​of all initial data segments in each monitoring data segment and the change characteristic values ​​of all initial data segments in the monitoring time series data.

[0092] In each monitoring time series data, the mean of the change characteristic values ​​of all initial data segments is used as the first characteristic factor, which represents the average level of data changes in the overall monitoring time series data; then, in each monitoring data segment, the mean of the change characteristic values ​​of all initial data segments is calculated as the second characteristic factor, which represents the average level of data changes in each monitoring data segment.

[0093] Finally, the absolute value of the difference between the second characteristic factor of each monitoring data segment and the corresponding first characteristic factor is taken as the degree of change value of each monitoring data segment, denoted as BC. The degree of change value represents the deviation between the change of data in each monitoring data segment and the change of the overall monitoring time series data. The larger the value, the greater the deviation, which means that the data change in the monitoring data segment is more complicated than the overall monitoring time series data, and has more prominent change characteristics.

[0094] Step S212: Determine the comparison data segment of each monitoring data segment in each monitoring time series data in other monitoring time series data.

[0095] In order to analyze the relationship between different monitoring time series data, it is necessary to find the corresponding or similar parts in the data. At the same time, since the interaction relationship between different monitoring time series data will not have a large time delay, in each monitoring data segment of each monitoring time series data, the mean of all data values ​​at the moment is used as the reference value of each monitoring data segment. The reference value can provide a stable reference point for matching the monitoring data segments in different monitoring time series data.

[0096] Finally, any monitoring data segment is selected as the target data segment in any type of monitoring time series data, and in each other monitoring time series data, the monitoring data segment with the smallest absolute difference with the reference value of the target data segment is used as the comparison data segment of the target data segment in each other monitoring time series data. Through this method, for each monitoring data segment in each monitoring time series data, the corresponding comparison data segment can be found in other monitoring time series data.

[0097] Step S213: In each monitoring time series data, the influence coefficient of each monitoring time series data is calculated by comprehensively considering the similarity, change degree value and average diffusion rate of the monitoring data segment and its corresponding comparison data segment.

[0098] The dynamic time warping algorithm can be used to calculate the similarity between two data segments, even if they are not completely aligned on the time axis, it can still capture the similarity in shape or trend. Therefore, in each monitoring time series data, the DTW value between each monitoring data segment and all corresponding comparison data segments is combined to determine the first influencing factor of each monitoring data segment.

[0099] Then, the absolute value of the difference between the change degree values ​​of each monitoring data segment and each corresponding comparison data segment is taken as the change difference factor, and the change difference factors between each monitoring data segment and all corresponding comparison data segments are accumulated to obtain the second impact factor of each monitoring data segment.

[0100] According to the first influencing factor, the second influencing factor and the average diffusion rate, the influencing characteristic value of each monitoring data segment is obtained. The formula model of the influencing characteristic value may be, for example:

[0101]

[0102] Among them, YT represents the impact characteristic value of each monitoring data segment; v represents the average diffusion rate in the brain magnetic resonance imaging results; BC represents the change degree value of each monitoring data segment; BC i represents the change value of the ith comparison data segment of each monitoring data segment; I represents the number of comparison data segments of each monitoring data segment; DTW i represents the DTW value between the i-th comparison data segment of each monitoring data segment; γ2 represents the preset second parameter.

[0103] In the formula model that affects the characteristic value, the sum of the DTW values ​​of each monitoring data segment and all corresponding comparison data segments is taken as the first influencing factor Since the smaller the DTW value, the more similar the two sets of data are, the smaller the value of the first impact factor, the higher the similarity between the monitoring data segments, reflecting the greater degree of mutual influence, so it is used as the denominator to achieve logical relationship correction; since the change degree value of the monitoring data segment represents the change characteristics of the data change in the monitoring data segment compared with the overall monitoring time series data, the absolute value of the difference between the change degree value of each monitoring data segment and each corresponding comparison data segment is taken as the change difference factor |BC-BC i |, the larger the value, the greater the difference in the change characteristics of the two data segments, otherwise they are considered to be more similar. Then, all the change difference factors corresponding to each monitoring data segment are accumulated to obtain the second influencing factor The smaller the second impact factor is, the more similar the change characteristics between each monitoring data segment and its corresponding comparison data segment are, which reflects the higher degree of mutual influence, so it is used as the denominator. At the same time, considering that the average diffusion rate v in the brain magnetic resonance imaging results can reflect the current degree of damage to the brain tissue, it is placed as a weight in the numerator and involved in the calculation process of the impact eigenvalue, thereby obtaining the impact eigenvalue of each monitoring data segment. The larger the impact eigenvalue is, the higher the degree of influence of the monitoring data segment on its corresponding comparison data segment.

[0104] Finally, in each monitoring time series data, the influencing characteristic values ​​of all monitoring data segments are integrated, and the sum of the influencing characteristic values ​​of all monitoring data segments is normalized to obtain the value as the influencing coefficient of each monitoring time series data, denoted as YX, with a value range of 0 to 1. The larger the value, the greater the degree of influence of this type of monitoring time series data on other monitoring time series data.

[0105] It should be noted that the normalization process can adopt the norm() normalization function, and the specific normalization method is not limited here; the method for obtaining the DTW value is a process well known to those skilled in the art and will not be described in detail here; the role of the preset second parameter γ2 is to prevent the denominator from being 0, and the value can be 0.001. The specific value can be adjusted according to the implementation scenario and is not limited here.

[0106] At this point, the influence coefficient of each monitoring time series data can be determined for use in subsequent analysis.

[0107] Step S3: In all monitoring time series data, based on the data change trend of the monitoring data segment at each moment and the influence coefficient of each monitoring time series data, determine the cognitive ability decline index of the subject to be tested at each moment; based on the difference between the cognitive ability decline index of the subject to be tested at different moments and the interval between the moments, calculate the intervention index required for the subject to be tested at each moment.

[0108] Since the changing trend of the values ​​in the monitoring time series data can reflect the state of cognitive ability of the stroke subjects, the cognitive ability decline index of the subjects at each specific moment is comprehensively calculated by analyzing the data change trend in all monitoring time series data and combining the influence coefficient of each monitoring time series data. On the basis of determining the cognitive ability decline index, further analyzing the differences between these indexes at different moments and the time intervals between these moments can infer the changes in cognitive ability decline. With this information, the intervention index that the subjects need to intervene at each moment can be calculated. The intervention index directly reflects the urgency and importance of intervention at a certain point in time, and provides an effective basis for subsequent doctors to make judgments and formulate personalized intervention plans.

[0109] Preferably, in one embodiment of the present invention, the method for obtaining the cognitive ability decline index includes:

[0110] See also Figure 4 , which shows a method flow chart of a method for obtaining a cognitive ability decline index in one embodiment of the present invention, the method comprising the following steps:

[0111] Step S301: In each monitoring time series data, the cognitive ability decline factor of each monitoring data segment is determined based on the data change trend of the monitoring data segment.

[0112] In step S202, the trend characteristic value of the initial data segment is calculated, which characterizes the change trend of the data values ​​in each initial data segment. If the value is positive and the larger it is, the overall change trend is increasing. Conversely, if the value is negative and the smaller it is, the overall change trend is decreasing.

[0113] Because the trend change of the data can reflect the change of the cognitive ability of the subject to be tested to a certain extent, in each monitoring data segment, the initial data segment with a positive trend characteristic value is taken as an ascending data segment, and the initial data segment with a negative trend characteristic value is taken as a descending data segment.

[0114] In each rising data segment, the difference between the data values ​​​​between the right endpoint and the left endpoint is taken as the rising factor, and the sum of the rising factors of all rising data segments is taken as the rising characteristic value; in each falling data segment, the difference between the data values ​​​​between the left endpoint and the right endpoint is taken as the falling factor, and the sum of the falling factors of all falling data segments is taken as the falling characteristic value; the rising characteristic value and the falling characteristic value more intuitively represent the changes in the entire segment of data.

[0115] In each monitoring data segment, the ratio between the rising characteristic value and the falling characteristic value is used as the rising relative value of each monitoring data segment. The rising relative value represents the relative situation of the rising degree and the falling degree of the data in the monitoring data segment. The larger the value, the more significant the rising trend of the data. The ratio of the number of rising data segments to the total number of initial data segments is used as the quantity factor. The larger the quantity factor, the more initial data segments in the rising trend are in each monitoring data segment. The product of the rising relative value and the quantity factor is used as the rising coefficient of each monitoring data segment. The larger the rising coefficient, the more significant the change of data in the monitoring data segment is in the rising trend.

[0116] In view of the different relationships between the changing trends of different types of detection time series data and the cognitive ability decline of the subjects to be tested, taking the monitoring time series data in the embodiment of the present invention as an example, the higher the glycated hemoglobin level, usually means that the individual has a higher blood sugar level, and in the field of neuroscience, high blood sugar state is believed to lead to decreased cognitive function and increased risk of dementia; the higher the MoCA scale score and the MMSE scale score, the better the individual's cognitive function, and the lower the score, the more likely it is that the individual has impaired or declining cognitive function.

[0117] Therefore, it is necessary to determine the relationship between each monitoring time series data and cognitive decline. If it is a positive correlation, such as the glycated hemoglobin level in the embodiment of the present invention, the normalized value of the rising coefficient of the monitoring data segment is used as the cognitive decline factor; if it is a negative correlation, such as the MoCA scale score and the MMSE scale score in the embodiment of the present invention, the normalized value of the reciprocal of the rising coefficient of the monitoring data segment is used as the cognitive decline factor, and the cognitive decline factor is recorded as SY; the larger the cognitive decline factor, the higher the possibility that the subject to be tested will experience cognitive decline in the corresponding period of the monitoring data segment.

[0118] It should be noted that the normalization method may use the norm() normalization function, and the specific normalization method is not limited here.

[0119] Step S302: In each monitoring time series data, the cognitive ability decline characteristic value of each monitoring data segment is determined according to the difference between the cognitive ability decline factors of the monitoring data segments.

[0120] In each monitoring time series data, since the first monitoring data segment can represent the first time period in which the subject to be tested experiences cognitive decline, the difference in cognitive decline factors between each monitoring data segment and the first monitoring data segment can be analyzed to determine the cognitive decline characteristic value. Specifically, the cognitive decline factor of each monitoring data segment is used as the numerator, and the sum of the cognitive decline factor of the first monitoring data segment and the preset third parameter γ3 is used as the denominator. The resulting ratio is used as the cognitive decline characteristic value, which can be regarded as the change in cognitive ability of each monitoring data segment relative to the first monitoring data segment. The larger the cognitive decline characteristic value is, the higher the possibility of cognitive decline in the subject to be tested in the subsequent process. On the contrary, the smaller the cognitive decline characteristic value is, the lower the possibility of cognitive decline in the subject to be tested in the subsequent process, and the better the recovery.

[0121] It should be noted that the purpose of presetting the third parameter γ3 is to prevent the denominator from being 0, and the value may be 0.001. The specific value may be adjusted according to the implementation scenario and is not limited here.

[0122] Step S303: weighted adjustment is performed on the cognitive ability decline characteristic value of the monitoring data segment based on the influence coefficient corresponding to the monitoring time series data, so as to determine the cognitive ability decline index of the subject to be tested at each moment.

[0123] Based on the above steps, the cognitive ability decline characteristic value corresponding to each monitoring data segment can be obtained in each monitoring time series data. In order to characterize the dynamic changes of the cognitive ability decline of the subject to be tested over a period of time, it is necessary to obtain the cognitive ability decline index at each moment.

[0124] Since the influence coefficient of monitoring time series data represents its influence on other monitoring time series data, the larger the value, the greater the influence of this type of monitoring time series data on other monitoring time series data, which can be regarded as a greater contribution to the analysis of the cognitive ability decline of the subject to be tested. Therefore, the influence coefficient corresponding to the type of monitoring time series data is used to perform weighted fusion on the cognitive ability decline characteristic value of the monitoring data segment, so as to determine the cognitive ability decline index of each subject to be tested at each moment.

[0125] First, the product of the cognitive ability decline characteristic value of each monitoring data segment and the influence coefficient corresponding to the monitoring time series data is used as the cognitive ability decline coefficient of each monitoring data segment, and weighting is achieved through this method. Then, the sum of the cognitive ability decline coefficients of the monitoring data segment in all types of monitoring time series data at each moment is normalized and used as the cognitive ability decline index of the subject to be tested at each moment, recorded as SZ, so as to achieve fusion. The larger the cognitive ability decline index obtained in the end, the more serious the cognitive decline of the subject to be tested at that moment.

[0126] It should be noted that the normalization method may adopt the norm() normalization function, and the specific normalization method is not limited here.

[0127] At this point, the cognitive ability decline index of the subject at each moment can be obtained.

[0128] Based on the determination of the cognitive ability decline index, we can further analyze the intervention index that the test subject needs at each moment. The level of the intervention index reflects the urgency and importance of the intervention treatment required by the test subject at a certain moment, and can provide data support for subsequent doctors to formulate treatment plans for the test subject.

[0129] Preferably, in one embodiment of the present invention, the method for obtaining the intervention index includes:

[0130] The time interval between each moment and the first moment was taken as the moment difference value; the normalized value of the difference between the cognitive ability decline index at the first moment and each moment was taken as the cognitive recovery index.

[0131] According to the cognitive ability decline index, moment difference value and cognitive recovery index corresponding to each moment, the intervention index that the subject needs to intervene at each moment is determined, and the cognitive ability decline index is positively correlated with the intervention index, and the moment difference value and cognitive recovery index are negatively correlated with the intervention index. The formula model of the intervention index can be specifically, for example:

[0132]

[0133] Among them, GZ represents the intervention index at each moment; SZ represents the cognitive ability decline index at each moment; Δt represents the moment difference value corresponding to each moment; SZ1 represents the cognitive ability decline index at the first moment; γ4 represents the preset fourth parameter; Sigmoid() represents the normalization function; norm() represents the normalization function.

[0134] In the formula model of the intervention index, when the cognitive ability decline index of the subject to be tested is higher, the degree of intervention on the subject to be tested should be increased accordingly, so the cognitive ability decline index at each moment is used as the numerator; when the time interval between each moment and the first moment is smaller, that is, the smaller the moment difference value is, the closer it is to the stroke occurrence moment, and the closer it is to the optimal treatment time, so the degree of intervention should be increased accordingly, so it is placed in the denominator; then calculate the difference SZ1-SZ between the cognitive ability decline index at the first moment and each moment. If the value is positive and the larger it is, it is considered that the cognitive ability of the subject to be tested has been restored, then the intervention index can be appropriately reduced. On the contrary, if the value is negative and the smaller it is, it means that the degree of cognitive ability decline of the subject to be tested may be aggravated, then the intervention index should be appropriately increased. Since the value may be positive or negative, it is normalized using Sigmoid() to obtain the cognitive recovery index Sigmoid(SZ1-SZ). The larger the value, the lower the intervention index should be, and the smaller the value, the higher the intervention index should be, so it is placed in the denominator. Finally, the obtained ratio is normalized and used as the intervention index for the object to be tested at each moment.

[0135] It should be noted that the purpose of presetting the fourth parameter γ4 is to prevent the denominator from being 0, and the value may be 0.001. The specific value may be adjusted according to the implementation scenario and is not limited here.

[0136] Step S4: Based on the cognitive ability decline index of the subject at all times and the intervention index requiring intervention, an auxiliary assessment is performed on the cognitive impairment risk of the subject.

[0137] Through the above steps, the cognitive ability decline index of the stroke subject at each moment and the intervention index required for intervention can be obtained. By drawing a line graph of the cognitive ability decline index and a line graph of the intervention index, it is helpful for doctors to more intuitively obtain the dynamic changes of the cognitive ability of the stroke subject over a period of time and the required intervention index, which is convenient for doctors to make judgments and formulate treatment plans or rehabilitation plans for patients.

[0138] It is also possible to set a cognitive decline threshold and an intervention threshold, and count the number of moments when the cognitive decline threshold is greater than the intervention threshold and the intervention index is greater than the intervention threshold. If the ratio of this number to the total number of moments is greater than a preset ratio threshold, an early warning is issued to help doctors make further judgments based on the above line graph.

[0139] It should be noted that the cognitive decline threshold and intervention threshold are both set at 0.6, and the ratio threshold is set at 0.65. The specific values ​​can be adjusted according to the implementation scenario and are not limited here.

[0140] In order to facilitate calculation, all indicator data involved in the calculation in the embodiment of the present invention are preprocessed to eliminate the dimension effect. The specific means of eliminating the dimension effect are technical means well known to those skilled in the art and are not limited here.

[0141] In summary, by obtaining a variety of monitoring time series data of the stroke subject to be tested and the average diffusion rate in the brain magnetic resonance imaging results, and combining these data for analysis, the cognitive function status of the subject to be tested can be more comprehensively reflected, thereby improving the accuracy of the auxiliary assessment of cognitive impairment risk. In the present invention, adaptive segmentation is performed based on the change of data values ​​in each monitoring time series data to obtain monitoring data segments; at the same time, the similarities between the monitoring data segments in different monitoring time series data are analyzed, and the influence coefficient of each monitoring time series data is determined in combination with the average diffusion rate. This step makes the evaluation process more personalized and can better adapt to the dynamic changes in the data and the differences between different individuals. Based on the data change trend and influence coefficient of the monitoring data segment, the cognitive ability decline index of the subject to be tested at each moment is determined, and the intervention index that needs to be intervened at each moment is calculated based on the cognitive ability decline index. This dynamic evaluation method can more accurately reflect the development and change of the patient's cognitive ability, provide doctors with more accurate auxiliary evaluation data, and enable doctors to dynamically monitor the cognitive decline of the subject to be tested in the subsequent process, timely discover and take intervention measures, so as to formulate more personalized treatment plans and rehabilitation plans for the subject to be tested, and improve the treatment effect and quality of life.

[0142] The present invention also proposes a stroke cognitive impairment risk auxiliary assessment system, please refer to Figure 5 , which shows a system block diagram, including: a data acquisition module 501, used to implement step S1 in the above method; an influence coefficient analysis module 502, used to implement step S2 in the above method; an indicator calculation module 503, used to implement step S3 in the above method; an auxiliary evaluation module 504, used to implement step S4 in the above method.

[0143] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the risk auxiliary assessment system for cognitive impairment of stroke provided in the above embodiment and the risk auxiliary assessment method for cognitive impairment of stroke are of the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0144] See also Figure 6 , which shows a system structure diagram of a stroke cognitive impairment risk auxiliary assessment system provided by an embodiment of the present invention, including a processor 600, a memory 601, a bus 602 and a communication interface 603, wherein the processor 600, the communication interface 603 and the memory 601 are connected via the bus 602; wherein the memory 601 may include a high-speed random access memory, the bus 602 may be an ISA bus, a PCI bus or an EISA bus, etc., and the processor 600 may be an integrated circuit chip with signal processing capabilities; the memory 601 stores at least one instruction, at least one program, a code set or an instruction set, and when the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by the processor, the steps in a stroke cognitive impairment risk auxiliary assessment method are implemented.

[0145] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0146] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for assisting in assessing the risk of cognitive impairment in stroke, characterized in that: The method comprises: Obtaining various monitoring time series data of stroke subjects and average diffusion rate in brain magnetic resonance imaging results; Based on the change of data values ​​in each monitoring time series data, each monitoring time series data is adaptively segmented to obtain multiple monitoring data segments in each monitoring time series data; analyzing the similarity of changes between data values ​​in monitoring data segments of different monitoring time series data, and determining the influence coefficient of each monitoring time series data in combination with the average diffusion rate; In all monitoring time series data, based on the data change trend of the monitoring data segment at each moment and the influence coefficient of each monitoring time series data, the cognitive ability decline index of the subject to be tested at each moment is determined; based on the difference between the cognitive ability decline index of the subject to be tested at different moments and the interval between moments, the intervention index of the subject to be tested that needs to be intervened at each moment is calculated; Based on the cognitive ability decline index of the subject at all times and the intervention index requiring intervention, an auxiliary assessment of the cognitive impairment risk of the subject is performed.

2. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 1, characterized in that: The method for acquiring the monitoring data segment includes: Each monitoring time series data is segmented based on an adaptive piecewise constant approximation algorithm to obtain an initial data segment; In each monitoring time series data, the variation range of the data value in each initial data segment is analyzed to obtain the variation characteristic value of each initial data segment; Obtain the curve corresponding to each initial data segment, and take the average of the tangent slope values ​​at all data values ​​in each initial data segment as the trend characteristic value of each initial data segment; For any two adjacent initial data segments in each monitoring time series data, the absolute value of the difference between the change characteristic values ​​is used as the change difference factor, and the absolute value of the sum of the trend characteristic values ​​is used as the trend similarity factor. According to the change difference factor and the trend similarity factor, the merging index between the two adjacent initial data segments is calculated, and the change difference factor is negatively correlated with the merging index, and the trend similarity factor is positively correlated with the merging index; In each monitoring time series data, the first initial data segment is taken as the data segment to be tested to start traversal, and it is determined whether the merging index between the data segment to be tested and the next adjacent initial data segment meets the preset condition. If not, the current data segment to be tested is taken as a monitoring data segment, and the next adjacent initial data segment is taken as a new data segment to be tested to continue traversal; If the condition is satisfied, the next adjacent initial data segment is merged into the data segment to be tested to obtain a new data segment to be tested, and it is determined whether the merging index between the last initial data segment in the new data segment to be tested and the next adjacent initial data segment meets the preset condition. If not, the current data segment to be tested is used as a detection data segment, and the next adjacent initial data segment is used as a new data segment to be tested to continue traversing; After the traversal is completed, all monitoring data segments of each monitoring time series data are obtained; The preset condition is that the merging index is greater than or equal to a preset merging threshold.

3. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 2, characterized in that: The method for obtaining the change characteristic value includes: In each initial data segment, the difference between the maximum data value and the mean of all data values ​​is taken as the first fluctuation factor of each initial data segment, the variance of all data values ​​is taken as the second fluctuation factor, and the sum of the first fluctuation factor and the second fluctuation factor is taken as the change characteristic value of each initial data segment.

4. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 3, characterized in that: The method for obtaining the influence coefficient includes: In each monitoring time series data, based on the difference between the change characteristic value of the initial data segment in each monitoring data segment and the change characteristic values ​​of all initial data segments in each monitoring time series data, determine the change degree value of each monitoring data segment; Determine the comparison data segment of each monitoring data segment in each monitoring time series data in other monitoring time series data; In each monitoring time series data, the DTW values ​​between each monitoring data segment and all corresponding comparison data segments are comprehensively analyzed to determine the first impact factor of each monitoring data segment; the difference between the change degree values ​​of each monitoring data segment and all corresponding comparison data segments is comprehensively analyzed to obtain the second impact factor of each monitoring data segment; According to the first influencing factor, the second influencing factor and the average diffusion rate, an influencing characteristic value of each monitoring data segment is obtained, wherein the first influencing factor and the second influencing factor are both negatively correlated with the influencing characteristic value, and the average diffusion rate is positively correlated with the influencing characteristic value; In each monitoring time series data, the sum of the influencing characteristic values ​​of all monitoring data segments is normalized and used as the influencing coefficient of each monitoring time series data.

5. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 4, characterized in that: The method for obtaining the comparison data segment includes: In each monitoring data segment of each monitoring time series data, the mean value of all data values ​​at corresponding moments is used as the reference value of each monitoring data segment; In any type of monitoring time series data, any monitoring data segment is selected as the target data segment. In each other monitoring time series data, the monitoring data segment with the smallest absolute difference with the reference value of the target data segment is selected as the comparison data segment of the target data segment in each other monitoring time series data.

6. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 1, characterized in that: The types of monitoring time series data include at least glycosylated hemoglobin level, MoCA scale score and MMSE scale score.

7. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 2, characterized in that: The method for obtaining the cognitive ability decline index comprises: In each monitoring time series data, the cognitive ability decline factor of each monitoring data segment is determined based on the data change trend of the monitoring data segment; In each monitoring time series data, the cognitive ability decline characteristic value is determined according to the difference between the cognitive ability decline factor of each monitoring data segment and the cognitive ability decline factor of the first monitoring data segment; The product of the cognitive ability decline characteristic value of each monitoring data segment and the corresponding influence coefficient is taken as the cognitive ability decline coefficient of each monitoring data segment; the sum of the cognitive ability decline coefficients of the monitoring data segment in all types of monitoring time series data at each moment is normalized and used as the cognitive ability decline index of the subject to be tested at each moment.

8. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 7, characterized in that: The method for obtaining the cognitive ability decline factor comprises: Among all the initial data segments, the initial data segments with positive trend characteristic values ​​are regarded as rising data segments, and the initial data segments with negative trend characteristic values ​​are regarded as falling data segments; In each rising data segment, the difference between the data values ​​of the right endpoint and the left endpoint is used as the rising factor, and the sum of the rising factors of all rising data segments is used as the rising characteristic value; In each descending data segment, the difference between the data values ​​of the left endpoint and the right endpoint is used as the descending factor, and the sum of the descending factors of all descending data segments is used as the descending characteristic value; In each monitoring data segment, the ratio between the rising characteristic value and the falling characteristic value is used as the rising relative value of each monitoring data segment; the ratio between the number of rising data segments and the total number of initial data segments in each monitoring data segment is used as the quantity factor; the product of the rising relative value and the quantity factor is used as the rising coefficient of each monitoring data segment; Determine the relationship between each monitoring time series data and cognitive decline. If it is a positive correlation, the normalized value of the rising coefficient of the monitoring data segment is used as the cognitive decline factor. If it is a negative correlation, the normalized value of the inverse of the rising coefficient of the monitoring data segment is used as the cognitive decline factor.

9. A method for assisting in assessing the risk of cognitive impairment in stroke according to claim 1, characterized in that: The method for obtaining the intervention index includes: The time interval between each moment and the first moment is taken as the moment difference value; the normalized value of the difference between the cognitive ability decline index at the first moment and each moment is taken as the cognitive recovery index; According to the cognitive ability decline index, moment difference value and cognitive recovery index corresponding to each moment, the intervention index required for the subject to be tested at each moment is determined, and the cognitive ability decline index is positively correlated with the intervention index, and the moment difference value and cognitive recovery index are negatively correlated with the intervention index.

10. A stroke cognitive impairment risk auxiliary assessment system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for auxiliary assessment of the risk of cognitive impairment in stroke as claimed in any one of claims 1 to 9 are implemented.

Images