Prediction method and system of cognitive impairment in neurology based on machine learning

Abstract

The present invention discloses a method and system for predicting cognitive impairment in neurology based on machine learning, including: unifying monitoring data of different biomarkers and normalizing the MoCA score of cognitive impairment. By fitting corrected values ​​of biomarker data and corrected values ​​of cognitive impairment scores, based on the biomarker monitoring data fitting curve and the MoCA score fitting curve, the present invention accurately calculates the degree to which biomarkers reflect cognitive impairment, and further assesses the risk of cognitive impairment in patients. This can effectively avoid the problem of inaccurate cognitive impairment prediction results caused by the influence of some biomarker data that has a low degree of reflection on the patient's cognitive impairment.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine learning and relates to a method and system for predicting cognitive impairment in neurology based on machine learning. Background Art

[0002] Cognitive impairment in neurology generally refers to a decline in various cognitive abilities related to brain function, affecting a patient's thinking, memory, language, judgment, and daily living. This impairment can be caused by a variety of neurological diseases, such as Alzheimer's disease, vascular dementia, and Lewy body dementia. Before these neurological diseases manifest overt symptoms, patients often experience physiological and cognitive changes. Therefore, predicting cognitive impairment can help doctors identify potential risks early, thereby improving patients' quality of life, slowing disease progression, and reducing social and family burdens.

[0003] Existing methods for predicting cognitive impairment mainly rely on cognitive function assessment tools (such as the Mini-Mental State Examination (MMSE), the Montreal Cognitive Assessment (MoCA), etc.), neuroimaging examinations (such as MRI (magnetic resonance imaging) and CT (computed tomography), etc.), and biomarker testing. Each single prediction method may be affected by other factors, resulting in inaccurate prediction results. For example, predictions using cognitive function assessment tools are highly subjective, and biomarker testing alone may be affected by other diseases, causing some biomarker data to be affected by both cognitive impairment and other diseases, thereby reducing the accuracy of reflecting the patient's cognitive impairment. Therefore, traditional methods of using machine learning to predict a patient's cognitive impairment through biomarkers often ignore the different accuracy of different biomarkers in reflecting cognitive impairment, and thus cannot effectively obtain accurate prediction results. Summary of the Invention

[0004] The purpose of the present invention is to solve the problem in the prior art that different biomarkers have different accuracy in reflecting cognitive impairment, resulting in the inability to obtain accurate prediction results, and to provide a method and system for predicting cognitive impairment in neurology based on machine learning.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] Machine learning-based prediction methods for cognitive impairment in neurology include:

[0007] Collect biomarker data from patients' metabolites and perform MoCA scores on patients for cognitive impairment;

[0008] Unify the different biomarker monitoring data of patients and obtain different biomarker data correction values;

[0009] Normalize the cognitive impairment score data to obtain the corrected value of the cognitive impairment MoCA score;

[0010] Fitting different biomarker data correction values ​​and cognitive impairment score correction values ​​to obtain biomarker monitoring data fitting curves and cognitive impairment MoCA score fitting curves;

[0011] Based on the fitting curve of the patient's biomarker monitoring data and the fitting curve of the cognitive impairment MoCA score, the degree to which each patient's biomarker reflects their cognitive impairment is calculated;

[0012] Based on the deviation of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment, the patient's risk of cognitive impairment is obtained.

[0013] A further improvement of the present invention is:

[0014] Furthermore, the different biomarker monitoring data of the patient are unified to obtain different biomarker data correction values. Specifically, since the fluctuation ranges of different biomarker data are different and the corresponding fluctuation amplitudes are different, the different biomarker monitoring data are unified into the same fluctuation range:

[0015] ;

[0016] in, represents the unified data value of the i-th monitoring data of the u-th biomarker, represents the monitoring value of the i-th monitoring data of the u-th biomarker, represents the maximum monitored value in the u-th biomarker, represents the minimum monitored value in the u-th biomarker; represents the u-th biomarker data range, Indicates that the monitoring value of the u-th biological data is normalized so that the biomarker data Unified to within the range.

[0017] Furthermore, the cognitive impairment score data was normalized to obtain the corrected value of the cognitive impairment MoCA score, specifically:

[0018] ;

[0019] in, represents the correction value of the MoCA score of cognitive impairment of the i-th individual, represents the MoCA score of cognitive impairment i, represents the maximum value of the MoCA score for cognitive impairment, Indicates the minimum value of the MoCA score for cognitive impairment.

[0020] Furthermore, the fitting curves of biomarker monitoring data and MoCA score of cognitive impairment were obtained. Specifically, the time was used as the horizontal axis and the monitoring data was used as the vertical axis. The least square method was used to correct the cognitive impairment score data. and different biomarker data correction values Perform fitting to obtain the fitting curve.

[0021] Furthermore, based on the fitting curve of the patient's biomarker monitoring data and the fitting curve of the cognitive impairment MoCA score, the degree to which each patient's biomarker reflects their cognitive impairment is calculated, specifically:

[0022] The fitting curves of the biomarker monitoring data and the MoCA score for cognitive impairment were evenly divided into several segments according to their horizontal axes. Several data points were evenly marked on the fitting curve segments as representative points for analyzing the changing trends of the fitting curve segments.

[0023] Calculate the consistency between the slope of each representative point in the biomarker fitting curve segment and the slope of the corresponding representative point in the cognitive score fitting curve to obtain the correlation between the biomarker and cognitive impairment;

[0024] Based on the overall change trend of each biomarker fitting curve and the slope difference between adjacent representative points in each fitting curve, the possibility of each biomarker fitting curve being interfered with by other factors is calculated;

[0025] The degree of correlation between the biomarker and the cognitive score is obtained based on the correlation between each segment of the biomarker fitting curve and the cognitive score fitting curve and the possibility of interference.

[0026] Furthermore, the consistency of the slope corresponding to each representative point in the biomarker fitting curve segment and the slope of the corresponding representative point in the cognitive score fitting curve is calculated to obtain the correlation between the biomarker and the cognitive impairment condition, specifically:

[0027] ;

[0028] in, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, Indicates the number of representative points on the t-th segment fitting curve, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope of the i-th representative point on the t-th segment fitting curve of the cognitive score. Since when cognitive impairment increases, the biomarker may increase or decrease, that is, it may be positively correlated or negatively correlated. It represents the absolute value of the difference between the absolute values ​​of the slopes of the two curves at the same position. The smaller the absolute value of the difference, the closer the two curves are at this position, and the higher the correlation. The larger the absolute value, the more consistent the changes in the biomarker and cognitive score, indicating a higher correlation; To ensure ; To determine whether the biomarker data is positively or negatively correlated with the cognitive score; when the value is positive, indicates a positive correlation, otherwise it is a negative correlation.

[0029] Furthermore, based on the overall change trend of each biomarker fitting curve and the slope difference between adjacent representative points in each fitting curve, the possibility of each biomarker fitting curve being interfered with by other factors is calculated, specifically:

[0030] ;

[0031] ;

[0032] in, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. represents the number of representative points in the t-th fitting curve segment, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope difference between adjacent representative points in the t-th segment of the fitting curve. The larger the difference, the shorter the duration of the same change trend in the fitting curve segment, that is, the more likely it is interfered by other factors. represents the overall change trend of the t-th segment fitting curve of the u-th biomarker, It represents the difference in the overall change trend between the tth segment of the fitting curve and its adjacent fitting curve segments. The larger the difference, the more likely it is to be interfered by other factors. represents the biomarker data value corresponding to the last representative point on the t-th segment fitting curve of the u-th biomarker, Represents the biomarker data value corresponding to the first representative point on the t-th segment fitting curve of the u-th biomarker.

[0033] Furthermore, based on the correlation between each segment of the biomarker fitting curve and the cognitive score fitting curve and the possibility of interference, the degree of correlation between the biomarker and the cognitive score is obtained, specifically:

[0034] ;

[0035] in, represents the correlation between the u-th biomarker and cognitive score, represents the total number of fitting curve segments divided by the fitting curve of the u-th biomarker, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. The higher the possibility that the data in the fitting curve segment is interfered with, the less this segment of data can reflect the correlation between the biomarker and the cognitive score, so the smaller the proportion.

[0036] Furthermore, based on the deviation of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment, the risk level of the patient developing cognitive impairment is obtained, specifically:

[0037] ;

[0038] in, Indicates the patient's risk of cognitive impairment. represents the total number of biomarkers associated with cognitive impairment, represents the correlation between the u-th biomarker and cognitive score, represents the patient's u-th biomarker monitoring value, represents the standard value of the u-th biomarker; The higher the biomarker's response to the cognitive score, the greater the impact of the biomarker data on predicting the patient's risk of cognitive impairment. Furthermore, only when the correlation between the biomarker and the cognitive score is consistent with the deviation direction of the actual monitored value relative to the standard value can the data be considered to reflect an increased risk of cognitive impairment in the patient. Represents the normalization function.

[0039] The neurology cognitive impairment prediction system based on machine learning includes:

[0040] An acquisition module, which collects biomarker data from the patient's metabolites and performs a MoCA score on the patient's cognitive impairment;

[0041] a unification module, which unifies the patient's different biomarker monitoring data to obtain different biomarker data correction values;

[0042] a normalization processing module, which performs normalization processing on the cognitive impairment score data to obtain a corrected value of the cognitive impairment MoCA score;

[0043] A fitting module, wherein the fitting module fits different biomarker data correction values ​​and cognitive impairment score correction values ​​respectively to obtain a biomarker monitoring data fitting curve and a cognitive impairment MoCA score fitting curve;

[0044] a calculation module, wherein the calculation module calculates the degree to which each patient's biomarker reflects his or her cognitive impairment based on a fitting curve of the patient's biomarker monitoring data and a fitting curve of the cognitive impairment MoCA score;

[0045] An acquisition module is provided for acquiring the risk level of the patient developing cognitive impairment based on the deviation value of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention effectively integrates diverse information and improves data accuracy and comparability by unifying the monitoring data of different biomarkers and normalizing the MoCA score for cognitive impairment. By fitting the corrected values ​​of the biomarker data and the corrected values ​​of the cognitive impairment score, an intuitive fitting curve is formed, which helps to intuitively understand the relationship between biomarkers and cognitive impairment. Finally, based on the biomarker monitoring data fitting curve and the MoCA score fitting curve, the degree to which the biomarkers reflect cognitive impairment is accurately calculated, and then the risk of cognitive impairment in patients is assessed. This can effectively avoid the problem of inaccurate cognitive impairment prediction results caused by the influence of some biomarker data that has a low degree of reflection on the patient's cognitive impairment. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 Schematic diagram of the process of predicting neurological cognitive impairment based on machine learning of the present invention;

[0050] Figure 2 Schematic diagram of the structure of the neurological cognitive impairment prediction system based on machine learning of the present invention;

[0051] Figure 3Schematic diagram of representative points on the fitting curve. DETAILED DESCRIPTION

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0053] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0054] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0055] In the description of the embodiments of the present invention, it should be noted that if the terms "upper," "lower," "horizontal," "inner," etc. appear, the orientation or positional relationship indicated is based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the inventive product is typically placed when in use. These terms are merely for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. In addition, the terms "first," "second," etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0056] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0057] In the description of the embodiments of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0058] The present invention is described in further detail below with reference to the accompanying drawings:

[0059] See also Figure 1 The present invention discloses a method for predicting cognitive impairment in neurology based on machine learning, comprising:

[0060] S101, collect biomarker data from patients' metabolites and perform MoCA scores on patients' cognitive impairment;

[0061] S102, unifying different biomarker monitoring data of the patient to obtain different biomarker data correction values;

[0062] Since different biomarker data have different fluctuation ranges and corresponding fluctuation amplitudes, different biomarker monitoring data are unified into the same fluctuation range:

[0063] ;

[0064] in, represents the unified data value of the i-th monitoring data of the u-th biomarker, represents the monitoring value of the i-th monitoring data of the u-th biomarker, represents the maximum monitored value in the u-th biomarker, represents the minimum monitored value in the u-th biomarker; represents the u-th biomarker data range, Indicates that the monitoring value of the u-th biological data is normalized so that the biomarker data Unified to within the range.

[0065] S103, normalizing the cognitive impairment score data to obtain a corrected value of the cognitive impairment MoCA score;

[0066] ;

[0067] in, represents the correction value of the MoCA score of cognitive impairment of the i-th individual, represents the MoCA score of cognitive impairment i, represents the maximum value of the MoCA score for cognitive impairment, Indicates the minimum value of the MoCA score for cognitive impairment.

[0068] S104, fitting different biomarker data correction values ​​and cognitive impairment score correction values ​​respectively to obtain a biomarker monitoring data fitting curve and a cognitive impairment MoCA score fitting curve;

[0069] With time as the horizontal axis and monitoring data as the vertical axis, the least square method was used to correct the cognitive impairment score data. and different biomarker data correction values Perform fitting to obtain the fitting curve.

[0070] S105, calculating the extent to which each patient's biomarker reflects their cognitive impairment based on the patient's biomarker monitoring data fitting curve and the MoCA cognitive impairment score fitting curve;

[0071] S105.1. Divide the fitted curves of the biomarker monitoring data and the MoCA score into a number of segments evenly according to their horizontal axes; and evenly mark a number of data points on the fitted curve segments as representative points for analyzing the trend of changes in the fitted curve segments;

[0072] S105.2. Calculate the consistency of the slope corresponding to each representative point in the biomarker fitting curve segment with the slope of the corresponding representative point in the cognitive score fitting curve to obtain the correlation between the biomarker and cognitive impairment;

[0073] ;

[0074] in, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, Indicates the number of representative points on the t-th segment fitting curve, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope of the i-th representative point on the t-th segment fitting curve of the cognitive score. Since when cognitive impairment increases, the biomarker may increase or decrease, that is, it may be positively correlated or negatively correlated. It represents the absolute value of the difference between the absolute values ​​of the slopes of the two curves at the same position. The smaller the absolute value of the difference, the closer the two curves are at this position, and the higher the correlation. The larger the absolute value, the more consistent the changes in the biomarker and cognitive score, indicating a higher correlation; To ensure ; To determine whether the biomarker data is positively or negatively correlated with the cognitive score; when the value is positive, indicates a positive correlation, otherwise it is a negative correlation.

[0075] S105.3. Calculate the likelihood that each biomarker fitting curve segment is interfered with by other factors based on the overall change trend of each biomarker fitting curve segment and the slope difference between adjacent representative points in each fitting curve segment;

[0076] ;

[0077] ;

[0078] in, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. represents the number of representative points in the t-th fitting curve segment, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope difference between adjacent representative points in the t-th segment of the fitting curve. The larger the difference, the shorter the duration of the same change trend in the fitting curve segment, that is, the more likely it is interfered by other factors. represents the overall change trend of the t-th segment fitting curve of the u-th biomarker, It represents the difference in the overall change trend between the tth segment of the fitting curve and its adjacent fitting curve segments. The larger the difference, the more likely it is to be interfered by other factors. represents the biomarker data value corresponding to the last representative point on the t-th segment fitting curve of the u-th biomarker, Represents the biomarker data value corresponding to the first representative point on the t-th segment fitting curve of the u-th biomarker.

[0079] S105.4. Obtain the degree of correlation between the biomarker and the cognitive score based on the correlation between each segment of the biomarker fitting curve and the cognitive score fitting curve and the possibility of interference.

[0080] ;

[0081] in, represents the correlation between the u-th biomarker and cognitive score, represents the total number of fitting curve segments divided by the fitting curve of the u-th biomarker, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. The higher the possibility that the data in the fitting curve segment is interfered with, the less this segment of data can reflect the correlation between the biomarker and the cognitive score, so the smaller the proportion.

[0082] S106, based on the deviation of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment, obtain the patient's risk of cognitive impairment.

[0083] ;

[0084] in, Indicates the patient's risk of cognitive impairment. represents the total number of biomarkers associated with cognitive impairment, represents the correlation between the u-th biomarker and cognitive score, represents the patient's u-th biomarker monitoring value, represents the standard value of the u-th biomarker; The higher the biomarker's response to the cognitive score, the greater the impact of the biomarker data on predicting the patient's risk of cognitive impairment. Furthermore, only when the correlation between the biomarker and the cognitive score is consistent with the deviation direction of the actual monitored value relative to the standard value can the data be considered to reflect an increased risk of cognitive impairment in the patient. Represents the normalization function.

[0085] See also Figure 2 The present invention discloses a neurological cognitive impairment prediction system based on machine learning, comprising:

[0086] An acquisition module, which collects biomarker data from the patient's metabolites and performs a MoCA score on the patient's cognitive impairment;

[0087] a unification module, which unifies the patient's different biomarker monitoring data to obtain different biomarker data correction values;

[0088] a normalization processing module, which performs normalization processing on the cognitive impairment score data to obtain a corrected value of the cognitive impairment MoCA score;

[0089] A fitting module, wherein the fitting module fits different biomarker data correction values ​​and cognitive impairment score correction values ​​respectively to obtain a biomarker monitoring data fitting curve and a cognitive impairment MoCA score fitting curve;

[0090] a calculation module, wherein the calculation module calculates the degree to which each patient's biomarker reflects his or her cognitive impairment based on a fitting curve of the patient's biomarker monitoring data and a fitting curve of the cognitive impairment MoCA score;

[0091] An acquisition module is provided for acquiring the risk level of the patient developing cognitive impairment based on the deviation value of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment.

[0092] Example:

[0093] The present invention discloses a method for predicting cognitive impairment in neurology based on machine learning, comprising:

[0094] Step 1: Continuously monitor patient metabolite biomarker data and MoCA scores.

[0095] Data preparation: Existing studies have shown that changes in the levels of various metabolites in patients' urine and plasma are related to the patient's cognitive impairment. For example, the tryptophan level in the urine of patients with mild cognitive impairment will decrease, and the lysine level in the plasma will increase. Therefore, urine and blood tests are performed on patients every day, and the patient's metabolic biomarker data are obtained using liquid chromatography-mass spectrometry technology, nuclear magnetic resonance technology, and other methods and technologies. The monitoring is continuous for one month, and data from multiple patients are collected. The collected data are shown in Table 1. The number of patient samples selected in this invention is 100. The specific number of samples can be adjusted according to the actual situation of the hospital.

[0096] Table 1: Metabolic biomarker data of patients

[0097]

[0098] In order to analyze the correlation between patients' metabolites and cognitive function, patients need to take the MoCA cognitive assessment test every day and record their MoCA scores. The MoCA test covers multiple cognitive areas such as memory, attention, language, visual-spatial skills, executive function and abstract thinking. The total score of the MoCA is 30 points. Generally, a score below 26 is considered to be at risk of cognitive impairment.

[0099] Step 2: Based on the correlation between the patient's different biomarker data and cognitive score data, calculate the degree to which each biomarker reflects the patient's cognitive impairment.

[0100] Step 2.1: Obtain the change curve of each biomarker monitoring data and cognitive impairment score for each patient.

[0101] When a patient's biomarker data and cognitive impairment score change trends align or contradict each other, it indicates a correlation between the biomarker and the patient's cognitive impairment score. To avoid differences in the fluctuation ranges of different biomarker data and cognitive impairment score data, which could affect the determination of correlation, the different data sets must first be unified into the same data range. To analyze the changing trends of different biomarker data and cognitive impairment score data, after unifying the different data ranges, the data were fitted to obtain fitting curves for the different data sets.

[0102] This method integrates the correlation between biomarker data and cognitive impairment scores from multiple patients to determine the degree to which different biomarkers respond to a patient's cognitive impairment. This first requires determining the relationship between each patient's biomarker and cognitive impairment. The following example uses the analysis process of the correlation between a patient's biomarker and cognitive impairment as an example.

[0103] Since different biomarker data have different fluctuation ranges and corresponding fluctuation amplitudes, direct comparison of their changing trends is ineffective. Therefore, the different biomarker monitoring data are first unified into the same fluctuation range:

[0104] ;

[0105] in, represents the unified data value of the i-th monitoring data of the u-th biomarker, represents the monitoring value of the i-th monitoring data of the u-th biomarker, represents the maximum monitored value in the u-th biomarker, represents the minimum monitored value in the u-th biomarker. represents the u-th biomarker data range, Indicates that the monitoring value of the u-th biological data is normalized so that the biomarker data Unified to within the range.

[0106] The cognitive impairment score data ranges from 0 to 30, which is quite different from the biomarker data range. In order to facilitate the analysis of the correlation between the changing trends of biomarker data and cognitive impairment data, it is necessary to unify the cognitive impairment score data and the biomarker data range.

[0107] ;

[0108] in, represents the correction value of the MoCA score of cognitive impairment of the i-th individual, represents the MoCA score of cognitive impairment i, represents the maximum value of the MoCA score for cognitive impairment, Indicates the minimum value of the MoCA score for cognitive impairment.

[0109] In order to facilitate the analysis of the changing trends of biomarker data and cognitive impairment score data, the time (day) was used as the horizontal axis and the monitoring data was used as the vertical axis. The least squares method was used to correct the cognitive impairment score data. and different biomarker data correction values Perform fitting to obtain the fitting curve.

[0110] Step 2.2: Calculate the degree to which each patient's biomarker reflects their cognitive impairment based on the cognitive score curve and biomarker curve of each patient.

[0111] Based on the obtained cognitive impairment score fitting curve and the changing trends of the fitting curves of various biomarker data, the degree to which various biomarkers reflect the patient's cognitive impairment can be determined. The stronger the positive or negative correlation between the changing trends of the fitting curves, the more the corresponding biomarker can reflect the patient's cognitive impairment. Therefore, by calculating the correlation between each biomarker fitting curve and the cognitive impairment score fitting curve, the degree of biomarker response to cognitive impairment is obtained. However, factors such as the patient's diet and medication may also cause temporary changes in the patient's biomarker data, making it unable to effectively reflect the patient's risk of cognitive impairment. Therefore, it is necessary to reduce the influence weight of data suspected of being interfered with by other factors in the final calculation of the degree to which biomarkers reflect cognitive impairment, so as to avoid inaccurate cognitive impairment prediction results caused by other factors.

[0112] Because individual biomarker data may change temporarily due to the influence of patients' diet and medication, in order to reduce the impact of such data on the analysis of the correlation between the overall biomarker and cognitive impairment, the obtained fitting curve is evenly divided into 10 segments according to its horizontal axis. The degree to which the biomarker data in each small segment of the fitting curve reflects the cognitive impairment situation is analyzed, that is, the correlation between the corresponding fitting curves. The specific process is as follows (the following process is only performed in each small segment of the curve obtained after the fitting curve is segmented):

[0113] (1) Take cognitive score data as an example, Figure 3 As shown in Figure 2, 5 data points are evenly marked on the obtained fitting curve segment as representative points for analyzing the changing trend of the fitting curve segment.

[0114] (2) Calculate the consistency of the slope of each representative point in the biomarker fitting curve segment and the slope of the representative point in the cognitive score fitting curve. If the biomarker data is highly correlated with the cognitive score, then the degree of change of the biomarker relative to its range of change should be consistent with the degree of change of the cognitive score. Because the data obtained in step a has been normalized, the more consistent the change range of the biomarker fitting curve and the cognitive score fitting curve, the higher the correlation between the biomarker and the cognitive impairment. Therefore, calculate the consistency of the slope of each representative point in the biomarker fitting curve segment and the slope of the representative point in the cognitive score fitting curve:

[0115] ;

[0116] in, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, Indicates the number of representative points on the t-th segment fitting curve, here , represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope of the i-th representative point on the t-th segment fitting curve of the cognitive score. As cognitive impairment increases, biomarkers may increase or decrease, that is, they may be positively correlated or negatively correlated. It represents the absolute value of the difference between the absolute values ​​of the slopes of the two curves at the same position. The smaller the absolute value of the difference, the closer the two curves are at this position, and the higher the correlation. The larger the absolute value, the more consistent the changes in the biomarker and cognitive score, indicating a higher correlation. To ensure . It is used to determine whether the biomarker data is positively or negatively correlated with the cognitive score. When the value is positive, indicates a positive correlation, otherwise it is a negative correlation.

[0117] (3) Calculate the possibility that each segment of the biomarker fitting curve is interfered with by other factors. Changes in biomarker data caused by other interfering factors such as diet or drugs are usually short-lived and have a large difference in magnitude from the change in the absence of interfering factors. Therefore, based on the difference in slopes corresponding to different representative points of each fitting curve segment and the difference in overall change trends between adjacent fitting curve segments, calculate the possibility that each biomarker fitting curve segment is interfered with by other factors:

[0118] ;

[0119] ;

[0120] in, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. represents the number of representative points in the t-th fitting curve segment, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope difference between adjacent representative points in the t-th segment of the fitting curve. The larger the difference, the shorter the duration of the same change trend in the fitting curve segment, that is, the more likely it is to be interfered by other factors. represents the overall change trend of the t-th segment fitting curve of the u-th biomarker, It represents the difference in the overall change trend between the tth segment of the fitting curve and its adjacent fitting curve segments. The larger the difference, the more likely it is to be interfered by other factors. represents the biomarker data value corresponding to the last representative point on the t-th segment fitting curve of the u-th biomarker, Represents the biomarker data value corresponding to the first representative point on the t-th segment fitting curve of the u-th biomarker.

[0121] Therefore, according to the correlation between each segment of the biomarker fitting curve and the cognitive score fitting curve and the possibility of interference, the correlation degree between the biomarker and the cognitive score is calculated:

[0122] ;

[0123] in, represents the correlation between the u-th biomarker and cognitive score, Indicates the total number of fitting curve segments divided by the fitting curve of the u-th biomarker (here ), represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. The higher the possibility that the data in the fitting curve segment is interfered with, the less this segment of data can reflect the correlation between the biomarker and the cognitive score, so the smaller the proportion.

[0124] In step 2.3, the response degree of biomarkers to cognitive impairment in different patients is comprehensively analyzed to calculate the impact of cognitive impairment on each biomarker.

[0125] The correlation between a single patient's biomarker data and cognitive scores may be accidental due to the patient's own disease, so it is necessary to combine the correlation between different biomarker data and cognitive scores obtained from multiple patients to calculate the impact of cognitive impairment on each biomarker, that is, the degree to which each biomarker data responds to the patient's cognitive impairment. Here, the average value of the correlation between biomarkers and cognitive scores obtained from all patient data is used. , as the degree of impact of cognitive impairment on each biomarker.

[0126] Step 3: Predict the patient's risk of cognitive impairment based on the patient's biomarker data and the degree to which the biomarkers reflect cognitive impairment.

[0127] Each biomarker has its normal standard value. When a patient's biomarker content deviates from the normal value, the patient's risk of cognitive impairment is calculated based on the deviation of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment:

[0128] ;

[0129] in, Indicates the patient's risk of cognitive impairment. represents the total number of biomarkers associated with cognitive impairment, represents the correlation between the u-th biomarker and cognitive score, represents the patient's u-th biomarker monitoring value, represents the standard value of the u-th biomarker. The higher the degree of response of the biomarker to the cognitive score, the greater the impact of the biomarker data on predicting the risk of cognitive impairment in patients. When the correlation between the biomarker and the cognitive score is consistent with the deviation direction of the actual monitoring value relative to the standard value, the data can be considered to reflect the increased risk of cognitive impairment in patients (for example, when a biomarker of a patient is negatively correlated with the cognitive score, that is, When it is a negative number, must also be negative for the risk of cognitive impairment to increase). Represents the normalization function.

[0130] when When the patient is diagnosed with cognitive impairment, it is considered that the patient is at risk of developing cognitive impairment and needs timely treatment intervention.

[0131] An embodiment of the present invention provides a terminal device. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of each of the aforementioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of each module / unit in each of the aforementioned device embodiments are implemented.

[0132] The computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to accomplish the present invention.

[0133] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0134] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0135] The memory may be used to store the computer programs and / or modules, and the processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory.

[0136] If the module / unit integrated into the terminal device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0137] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A neurological cognitive impairment prediction system based on machine learning, characterized by: include: An acquisition module, which collects biomarker data from the patient's metabolites and performs a MoCA score on the patient's cognitive impairment; a unification module, which unifies the patient's different biomarker monitoring data to obtain different biomarker data correction values; a normalization processing module, which performs normalization processing on the cognitive impairment score data to obtain a corrected value of the cognitive impairment MoCA score; The fitting module fits different biomarker data correction values ​​and cognitive impairment score correction values ​​to obtain biomarker monitoring data fitting curves and cognitive impairment MoCA score fitting curves; specifically, the fitting module uses the least squares method to fit the cognitive impairment score correction values ​​using time as the horizontal axis and monitoring data as the vertical axis. and different biomarker data correction values Perform fitting to obtain a fitting curve; A calculation module is provided for calculating the degree to which each patient's biomarker reflects their cognitive impairment based on the patient's biomarker monitoring data fitting curve and the cognitive impairment MoCA score fitting curve; specifically: The fitting curves of the biomarker monitoring data and the MoCA score for cognitive impairment were evenly divided into several segments according to their horizontal axes. Several data points were evenly marked on the fitting curve segments as representative points for analyzing the changing trends of the fitting curve segments. Calculate the consistency between the slope of each representative point in the biomarker fitting curve segment and the slope of the corresponding representative point in the cognitive score fitting curve to obtain the correlation between the biomarker and cognitive impairment; specifically: ； in, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, Indicates the number of representative points on the t-th segment fitting curve, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope of the i-th representative point on the t-th segment fitting curve of the cognitive score. Since when cognitive impairment increases, the biomarker may increase or decrease, that is, it may be positively correlated or negatively correlated. It represents the absolute value of the difference between the absolute values ​​of the slopes of the two curves at the same position. The smaller the absolute value of the difference, the closer the two curves are at this position, and the higher the correlation. The larger the absolute value, the more consistent the changes in the biomarker and cognitive score, indicating a higher correlation; To ensure ; To determine whether the biomarker data is positively or negatively correlated with the cognitive score; when the value is positive, indicates positive correlation, otherwise negative correlation; Based on the overall change trend of each biomarker fitting curve and the slope difference between adjacent representative points in each fitting curve, the possibility of each biomarker fitting curve being interfered with by other factors is calculated; specifically: ； ； in, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. represents the number of representative points in the t-th fitting curve segment, represents the slope of the i-th representative point on the t-th segment fitting curve of the u-th biomarker, It represents the slope difference between adjacent representative points in the t-th segment of the fitting curve. The larger the difference, the shorter the duration of the same change trend in the fitting curve segment, that is, the more likely it is interfered by other factors. represents the overall change trend of the t-th segment fitting curve of the u-th biomarker, It represents the difference in the overall change trend between the tth segment of the fitting curve and its adjacent fitting curve segments. The larger the difference, the more likely it is to be interfered by other factors. represents the biomarker data value corresponding to the last representative point on the t-th segment fitting curve of the u-th biomarker, represents the biomarker data value corresponding to the first representative point on the t-th segment fitting curve of the u-th biomarker; According to the correlation between each segment of the biomarker fitting curve and the cognitive score fitting curve and the possibility of interference, the correlation degree between the biomarker and the cognitive score is obtained; An acquisition module is provided for acquiring the risk level of the patient developing cognitive impairment based on the deviation value of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment.

2. The neurological cognitive impairment prediction system based on machine learning according to claim 1, characterized in that: The method of unifying the different biomarker monitoring data of the patient to obtain different biomarker data correction values ​​is as follows: since different biomarker data have different fluctuation ranges and corresponding fluctuation amplitudes, the different biomarker monitoring data are unified into the same fluctuation range: ； in, represents the unified data value of the i-th monitoring data of the u-th biomarker, represents the monitoring value of the i-th monitoring data of the u-th biomarker, represents the maximum monitored value in the u-th biomarker, represents the minimum monitored value in the u-th biomarker; represents the u-th biomarker data range, Indicates that the monitoring value of the u-th biological data is normalized so that the biomarker data Unified to within the range.

3. The neurological cognitive impairment prediction system based on machine learning according to claim 2, characterized in that: The normalization process of the cognitive impairment score data is performed to obtain a corrected value of the cognitive impairment MoCA score, specifically: ； in, represents the correction value of the MoCA score of cognitive impairment of the i-th individual, represents the MoCA score of cognitive impairment i, represents the maximum value of the MoCA score for cognitive impairment, Indicates the minimum value of the MoCA score for cognitive impairment.

4. The neurological cognitive impairment prediction system based on machine learning according to claim 3, characterized in that: The correlation between each segment of the biomarker fitting curve and the cognitive score fitting curve and the possibility of interference are used to obtain the correlation between the biomarker and the cognitive score, specifically: ； in, represents the correlation between the u-th biomarker and cognitive score, represents the total number of fitting curve segments divided by the fitting curve of the u-th biomarker, represents the correlation between the t-th segment fitting curve of the u-th biomarker and the cognitive score, It indicates the possibility that the t-th fitting curve segment of the u-th biomarker is interfered by other factors. The higher the possibility that the data in the fitting curve segment is interfered with, the less this segment of data can reflect the correlation between the biomarker and the cognitive score, so the smaller the proportion.

5. The neurological cognitive impairment prediction system based on machine learning according to claim 4, characterized in that: The risk level of the patient developing cognitive impairment is obtained based on the deviation of the patient's biomarker data from its standard value and the degree to which the biomarker reflects cognitive impairment, specifically: ； in, Indicates the patient's risk of cognitive impairment. represents the total number of biomarkers associated with cognitive impairment, represents the correlation between the u-th biomarker and cognitive score, represents the patient's u-th biomarker monitoring value, represents the standard value of the u-th biomarker; The higher the biomarker's response to the cognitive score, the greater the impact of the biomarker data on predicting the patient's risk of cognitive impairment. Furthermore, only when the correlation between the biomarker and the cognitive score is consistent with the deviation direction of the actual monitored value relative to the standard value can the data be considered to reflect an increased risk of cognitive impairment in the patient. Represents the normalization function.

Images