Risk prediction method and system for type 2 diabetes mellitus complicated with mild cognitive impairment

Abstract

The invention discloses a risk prediction method and system for type 2 diabetes mellitus complicated with mild cognitive impairment, and relates to the field of data processing. The method comprises the following steps: acquiring potential risk factor related data of type 2 diabetes subjects, and dividing the subjects into an MCI group and an NC group through cognitive evaluation; the potential risk factors comprise demographic characteristics and clinical characteristics; screening target factors from the potential risk factors according to the related data of the potential risk factors and the cognitive assessment result; and constructing a column diagram about the target factor, and predicting the quantitative risk of the cognitive impairment according to the column diagram. According to the method, the risk can be quickly quantified through the column diagram, and the risk can be systematically evaluated by integrating a plurality of target factors, so that the cognitive dysfunction risk of the type 2 diabetes mellitus crowd can be conveniently and accurately predicted.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a risk prediction method and system for type 2 diabetes mellitus complicated with mild cognitive impairment. Background Art

[0002] As a complication of type 2 diabetes mellitus (T2DM), cognitive impairment has received increasing attention due to its profound impact on diabetes management and overall quality of life. Existing clinical and epidemiological evidence shows that patients with type 2 diabetes are at an increased risk of developing mild cognitive impairment (MCI) and Alzheimer's disease (AD). Studies have shown that the risk of cognitive decline in diabetic patients is 2-3 times that of non-diabetic patients.

[0003] MCI is a transitional stage between normal cognitive aging and dementia and may be modifiable. Therefore, early detection of MCI in diabetic patients is beneficial to the recovery of cognitive function, or to take effective measures to delay cognitive decline, which is beneficial to the self-management of diabetes.

[0004] However, there are few studies on quantitatively assessing the risk of MCI in T2DM patients, and there is a lack of a relatively fast and systematic risk prediction method. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a risk prediction method and system for type 2 diabetes mellitus complicated with mild cognitive impairment that is convenient and systematic.

[0006] In a first aspect, the present application provides a risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment. The method includes:

[0007] Obtain data related to potential risk factors of type 2 diabetes subjects, and divide the subjects into an MCI group and an NC group through cognitive assessment; the potential risk factors include demographic characteristics and clinical characteristics;

[0008] According to the data related to potential risk factors and the results of cognitive assessment, screen target factors from the potential risk factors;

[0009] Construct a nomogram for the target factors, and predict the quantitative risk of cognitive impairment according to the nomogram.

[0010] In one embodiment, ethnic-specific factors are introduced to construct the nomogram matching different ethnic groups.

[0011] In one embodiment, dividing the subjects into an MCI group and an NC group through cognitive assessment includes:

[0012] Cognitive assessment is performed by one or more of the Montreal Cognitive Assessment, Mini-Mental State Examination, Clock Drawing Test, DST, VFT, TMTA, TMTB, AVLT-IR, AVLT-DR, and LMT to obtain a quantified cognitive assessment result;

[0013] The quantified cognitive assessment result is compared with a threshold, and the subjects are divided into an MCI group and an NC group according to the comparison result.

[0014] In one embodiment, screening target factors from potential risk factors according to potential risk factor-related data and cognitive assessment results includes:

[0015] Conduct an inter-group difference analysis using t-test, Mann-Whitney test or chi-square test to determine the differences corresponding to each potential risk factor between the MCI group and the NC group;

[0016] Conduct a non-parametric Spearman correlation analysis to determine the correlation between each potential risk factor and cognitive function;

[0017] Calculate VAI or CVAI according to potential risk factor-related data, and use a multivariate linear regression method to evaluate the relationship between VAI or CVAI and cognitive function;

[0018] Use univariate logistic regression and two-way stepwise logistic regression analysis to screen target factors.

[0019] In one embodiment, quantifying the risk of mild cognitive impairment according to a nomogram includes:

[0020] Obtain the collected data of the research object regarding the target factors;

[0021] Based on the collected data, determine the scores corresponding to each target factor by looking up the nomogram;

[0022] Add up the scores corresponding to each target factor to obtain a total score, and calculate the risk quantification of mild cognitive impairment according to the total score. The calculation formula is as follows:

[0023] Risk=-7.92e-07×points 3 +0.000274248×points 2 -0.022018993×points+0.605121118;

[0024] In the formula, points represents the total score, and Risk represents the quantified risk of mild cognitive impairment.

[0025] In a second aspect, the present application also provides a risk prediction system for type 2 diabetes mellitus complicated with mild cognitive impairment. The system includes:

[0026] An input module for acquiring the collected data of the target factor;

[0027] A processing module for predicting the quantitative risk of cognitive impairment according to the collected data in combination with the nomogram;

[0028] An output module for outputting the quantitative risk.

[0029] In one embodiment, the system includes an ethnic identification module; the processing module is further configured to match the corresponding nomogram according to the ethnicity identified by the ethnic identification module, and predict the quantitative risk based on the matched nomogram.

[0030] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps in the above-mentioned risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment are implemented.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps in the above-mentioned risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment are implemented.

[0032] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment are implemented.

[0033] The above-mentioned risk prediction method and system for type 2 diabetes mellitus complicated with mild cognitive impairment include: acquiring data related to potential risk factors of type 2 diabetes mellitus subjects, and classifying the subjects into the MCI group and the NC group through cognitive assessment; the potential risk factors include demographic characteristics and clinical characteristics; screening target factors from the potential risk factors according to the data related to the potential risk factors and the cognitive assessment results; constructing a nomogram about the target factors, and predicting the quantitative risk of cognitive impairment according to the nomogram. By comprehensively considering the risk relevance of various potential risk factors to cognitive impairment in the type 2 diabetes population, screening target factors, and then constructing a nomogram based on the target factors, the quantitative risk can be quickly obtained through the nomogram, and the risk can be systematically evaluated by integrating multiple target factors, realizing convenient and accurate risk prediction of cognitive impairment in the type 2 diabetes population. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic flow chart of a risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment in one embodiment;

[0035] Figure 2 Schematic diagram of a nomogram in one embodiment;

[0036] Figure 3 Schematic diagram of a calibration curve in one embodiment;

[0037] Figure 4 Schematic diagram of an ROC curve in one embodiment;

[0038] Figure 5 Schematic diagram of a decision curve in one embodiment. Detailed implementation manners

[0039] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0040] The embodiment of the present application provides a risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment, as Figure 1 shown, including the following steps:

[0041] Step 102, obtain data related to potential risk factors of type 2 diabetes subjects, and divide the subjects into the MCI group and the NC group through cognitive assessment; the potential risk factors include demographic characteristics and clinical characteristics.

[0042] Among them, the MCI group refers to type 2 diabetes subjects with mild cognitive impairment in cognitive assessment, and the NC group refers to type 2 diabetes subjects with normal cognitive assessment. The generally adopted method is the Montreal Cognitive Assessment (MoCA), and the assessment score ranges from 0 to 3012. MoCA score ≥ 26 is classified into the NC group, and MoCA score < 26 is classified into the MCI group. In the MoCA assessment, if the informed years of education are less than 12 years, an adjustment score needs to be added to reduce the potential bias related to the education level 13

[0043] Demographic characteristics include age, gender, height, weight, diabetes duration, education level, and hypertension prevalence. Clinical characteristics mainly refer to the results of fasting venous blood tests, including fasting plasma glucose (FPG), glycated hemoglobin (HbA1c), total cholesterol (TC), triglyceride (TG), high-density lipoprotein (HDL) and low-density lipoprotein (LDL) levels, as well as alanine aminotransferase (ALT) and aspartate aminotransferase (AST), blood urea nitrogen (BUN) concentration and creatinine (Cr).

[0044] Step 104, screen target factors from the potential risk factors according to the data related to the potential risk factors and the cognitive assessment results.

[0045] Screen out the key factors affecting cognitive function from potential risk factors as target factors. The screening methods that can be used in this embodiment include statistical methods, machine learning methods, expert experience methods, etc.

[0046] To explore the risk of cognitive decline in patients with type 2 diabetes, this embodiment compares the demographic characteristics, clinical characteristics, and cognitive assessment results between the MCI group and the NC group. Compared with the NC group, patients in the MCI group were mostly female, with significantly higher diabetes duration and waist circumference, and significantly lower education level and hypertension prevalence (P<0.05). However, there were no significant differences in BMI, HbA1c, FBG, TG, TC, HDL, LDL, Cr, BUN, ALT, and AST between the two groups (P>0.05). Due to the inclusion of patients with poor blood glucose control who required hospitalization during the trial, the differences in HbA1c and FBG did not reach statistical significance.

[0047] From the above comparison results, it can be seen that the target factors at least include gender, diabetes duration, waist circumference, education level, and hypertension prevalence.

[0048] When studying mild cognitive impairment in patients with type 2 diabetes, it is found that in addition to being strongly associated with the target factors composed of the above single indicators, it is also highly associated with the visceral fat index composed of multiple potential risk factors. The dysfunction of visceral adipose tissue is one of the pathogenesis of insulin resistance and type 2 diabetes. Therefore, in one embodiment, the visceral fat index is used as a target factor.

[0049] However, due to the differences in body fat distribution among different races, the commonly used visceral adiposity index (VAI) is mainly applicable to the populations in Europe and the United States and is not applicable to the yellow race. To overcome this limitation, the Chinese Visceral Adiposity Index (CVAI) is introduced in the preferred embodiment.

[0050] The calculation formula of VAI is: VAI (male) = WC / (39.68 + 1.88 × BMI) × TG / 1.03 × 1.31 / HDL; VAI (female) = WC / (36.58 + 1.89 × BMI) × TG / 0.81 × 1.52 / HDL.

[0051] The calculation formula of CVAI is as follows: CVAI (for men) = -267.93 + 0.68 × age + 0.03 × BMI + 4.00 × WC + 22.00 × lgTG - 16.32 × HDL; CVAI (for women) = -187.32 + 1.71 × age + 4.23 × BMI + 1.12 × WC + 39.76 × lgTG - 11.66 × HDL.

[0052] In the above formula, WC is the waist circumference; BMI is the body mass index, and its calculation formula is: BMI = body weight (kg) / [height (m)].

[0053] CVAI is a visceral fat assessment index determined using a binary linear logistic regression model. Studies have shown that CVAI is a visceral fat index suitable for Asians compared to traditional fat indices (such as WC, BMI, WHR, VAI, LAP, BMI, etc.).

[0054] When the research object is the Mongoloid race, CVAI is taken as one of the target factors; otherwise, VAI is taken as one of the target factors. In other embodiments, the ethnic group can be further refined and an appropriate visceral fat assessment calculation scheme can be adopted.

[0055] In addition to different visceral fat indices, there are also differences in the combination of target factors among different ethnic groups.

[0056] Step 106: Construct a nomogram for the target factors and predict the quantitative risk of cognitive impairment based on the nomogram.

[0057] After determining the target factors, a new scoring system including all target factors is constructed and presented in the form of a nomogram to predict the risk of cognitive impairment in type 2 diabetes patients.

[0058] All target factors are assigned numerical scores for quantification, and the scores of each target factor are added up using the Figure 2 shown nomogram to calculate the risk scores corresponding to each target factor. Then, based on the total score of all target factors, the risk of cognitive impairment in type 2 diabetes patients is predicted. The larger the total score, the greater the risk. Figure 2 The shown nomogram takes CVAI as one of the target factors. In other embodiments, a nomogram taking VAI as one of the target factors is also constructed, and a specific nomogram is called for risk assessment after confirming the patient's ethnic group.

[0059] In this embodiment, a risk prediction model is constructed through a nomogram to quantitatively evaluate the risk of cognitive impairment in type 2 diabetes patients, providing a more intuitive and faster assessment method for patients and clinicians.

[0060] In one embodiment, step 106 includes: obtaining the acquisition data of the research object regarding the target factors; based on the acquisition data, by looking up the nomogram, determining the scores corresponding to each target factor; adding up the scores corresponding to each target factor to obtain the total score, and quantitatively calculating the risk of mild cognitive impairment according to the total score. The calculation formula is as follows:

[0061] Risk=-7.92e-07×points 3 +0.000274248×points 2 -0.022018993×points+0.605121118;

[0062] In the formula, points represents the total score, and Risk represents the quantified risk of mild cognitive impairment.

[0063] Since there are specificities in the combination of target factors among different ethnic groups, in one embodiment, an ethnic-specific factor is introduced to construct nomograms matching different ethnic groups. When the research object is of the yellow race, a target factor combination scheme including CVAI and its corresponding nomogram are matched. When the research object is of other ethnic groups, another target factor combination scheme including VAI or other visceral fat indices and its corresponding nomogram are matched.

[0064] As Figure 2 shown, it is the nomogram obtained with the yellow race as the research object. In this figure, the target factors include age (Age), gender (Gender), education level (Education), diabetes duration (DM Duration), hypertension (HTN), and CVAI. After obtaining the acquisition data corresponding to the above target factors, querying the nomogram can calculate the scores corresponding to each target factor.

[0065] The contribution of age to the score is a linear equation, and its score calculation formula is: P1 = 1.291091214×age - 51.643648561. It shows that the older the age, the higher the risk.

[0066] Gender is a binary variable. When the gender is female, the score contribution is P2 = 15.59103; when the gender is male, the score contribution is P2 = 0. It shows that the risk of females is higher than that of males.

[0067] The education level is referenced by the number of years of education, and its score calculation formula is P3=-3.597124554×education + 79.136740193. It shows that the higher the education level, the lower the risk.

[0068] The score calculation formula for diabetes duration is P4 = 0.163916636×diabetes duration. It shows that the longer the diabetes duration, the higher the risk.

[0069] Hypertension is also a binary variable. If there is no hypertension, the score contribution is P5 = 0; if there is hypertension, the score contribution is P5 = 5.4481.

[0070] The scoring formula for CVAI is P6 = 0.285714286 × CVAI. This indicates that the higher the abdominal obesity index, the higher the risk.

[0071] Sum up the calculated score contributions P1 to P6 to obtain the total score points, and substitute points into the Risk calculation formula to calculate the final MCI risk prediction result.

[0072] In one embodiment, in step 102, the subjects are divided into an MCI group (mild cognitive impairment group) and an NC group (normal cognitive function group) through the highly sensitive Montreal Cognitive Assessment (MoCA), and multi-dimensional cognitive assessments are performed, including: Mini-Mental State Examination (MMSE), Clock Drawing Test (CDT), Digit Span Test (DST), Verbal Fluency Test (VFT), Trail Making Test A and B (TMT-A and TMT-B), Auditory Verbal Learning Test - Immediate Recall (AVLT-IR) and Delayed Recall (AVLT-DR), and Logical Memory Test (LMT).

[0073] The Montreal Cognitive Assessment Scale is used to evaluate the cognitive impairment of research participants, covering multiple cognitive domains such as attention, executive function, memory, language, visuospatial ability, abstract thinking, calculation ability, and orientation. In addition, cognitive function can also be evaluated through the Mini-Mental State Examination (MMSE). The Clock Drawing Test (CDT) is used for visual-spatial function analysis. DST, VFT, TMTA, and TMTB are used to evaluate executive function and information processing speed. AVLT-IR and AVLT-DR detect immediate and delayed memory functions, and LMT is used to measure episodic memory function.

[0074] In this embodiment, the Montreal Cognitive Assessment can be used alone for neuropsychological testing, or it can be combined with other tests to complement each other's deficiencies, give full play to their strengths, and more comprehensively and accurately evaluate the cognitive situation of patients.

[0075] In one embodiment, first, the MMSE is used as a preliminary screening tool to quickly evaluate the overall cognitive function of patients. The MMSE covers dimensions such as orientation, memory, attention, calculation ability, recall ability, and language ability. This test usually takes 5 to 10 minutes and can quickly provide a quantitative preliminary judgment of the patient's cognitive function. The specific score obtained can be used as a basic indicator for judging whether the patient is within the normal cognitive range and the approximate degree of cognitive impairment. Next, the MoCA is used to conduct a more refined assessment of the patient's cognitive dimensions. Different from the overall assessment of the MMSE, the MoCA focuses on the assessment of executive function, attention, abstract thinking, etc. The combination of the two can provide a more comprehensive analysis of cognitive function. For example, when the MMSE score is at the critical value and the MoCA score is significantly lower than the normal range, it may indicate that the patient has mild cognitive impairment, especially in the early stage of cognitive impairment. Subsequently, other cognitive assessment tools, such as CDT, DST, VFT, TMTA, TMTB, AVLT-IR, AVLT-DR, LMT, etc., are used to further quantify the impairment of specific cognitive functions. For example, when the score of the clock drawing test (CDT) is low while the MMSE and MoCA scores are relatively high, it may indicate that the patient's visual-spatial function and executive function are more prominently impaired. At this time, combining the evaluation results of each tool helps to more accurately determine the type and severity of cognitive impairment. Finally, combining the characteristics of various assessment scales, relative weights are assigned to the scores of each assessment method to establish a comprehensive assessment framework. For example, the MMSE and MoCA are used as basic assessment tools to conduct preliminary screening and refined analysis of the overall cognitive status respectively. Other assessment tools further supplement the overall assessment according to the cognitive dimensions they assess. For example, the CDT mainly assesses visual-spatial function, the LMT is used to assess episodic memory function, and the DST, VFT, TMTA, and TMTB mainly assess executive function. The AVLT-IR and AVLT-DR respectively assess immediate memory and delayed memory functions. The scores of all assessment tools will be integrated according to their roles in different cognitive dimensions to provide a more accurate assessment of cognitive impairment for patients.

[0076] In one embodiment, step 104 includes: performing an analysis of inter-group differences using a t-test, Mann-Whitney test, or chi-square test to determine the differences corresponding to each potential risk factor between the MCI group and the NC group; performing a non-parametric Spearman correlation analysis to determine the correlations between the potential risk factors; calculating the VAI or CVAI based on the relevant data of the potential risk factors, and using a multiple linear regression method to evaluate the relationship between the VAI or CVAI and cognitive function; using a univariate logistic regression and a two-way stepwise logistic regression analysis to screen for target factors.

[0077] This application uses SPSS and R software for statistical analysis. After obtaining the data related to potential risk factors, it is necessary to standardize the data first, which is expressed as the mean ± standard deviation (SD) or the median and interquartile range (IQR). For the differences between groups, t-tests for normally distributed data types, Mann-Whitney tests for non-normally distributed data types, and chi-square tests for categorical variables are used to obtain the differences and identify the potential risk factors with relatively large differences between the MCI group and the NC group. For the correlation between potential risk factors and cognitive function indicators, non-parametric Spearman correlation analysis is used to complete the screening of possible independent variables for subsequent regression analysis. Based on the previous differential analysis and correlation analysis, a basis can be provided for the selection of independent variables, and some independent variables are screened out. Then, a multivariate linear regression method is used to evaluate the relationship between the independent variables and cognitive function. The independent variables include VAI or CVAI. Finally, univariate and multivariate logistic regression analyses are used to screen the target factors of MCI. Univariate analysis can initially determine whether there is an association between each independent variable and the dependent variable, and multivariate analysis is based on univariate analysis, considering the interaction between independent variables to screen out independent risk factors.

[0078] During the process of multivariate linear regression analysis, the ideas for screening independent variables and constructing models can also provide a reference for logistic regression analysis. For example, through linear regression, it is found that certain independent variables have a significant impact on continuous dependent variables, and these independent variables may also play an important role in logistic regression for binary dependent variables.

[0079] When performing multivariate logistic regression analysis, selecting appropriate independent variables is the key, and the results of previous correlation analysis, differential analysis, and multivariate linear regression can help exclude some irrelevant or redundant variables, improving the accuracy and stability of the model.

[0080] In this embodiment, the dependent variable is whether an individual has mild cognitive impairment, and the independent variables include potential risk factors. Univariate analysis shows that age, gender, diabetes duration, VAI / CVAI, education level, and hypertension prevalence are related to MCI. The significant risk factors determined in univariate analysis are used as variables in the multivariate logistic regression model for analysis to determine the final target factors.

[0081] It should be noted that among these factors, stepwise multivariate regression analysis shows that CVAI independently affects the MCI risk in type 2 diabetes patients. Specifically, for every 1 unit increase in CVAI, the risk of MCI increases by 1.1% (OR = 1.011, 95% CI is 1.005 - 1.017) (p < 0.001). VAI also shows a similar independent influence characteristic.

[0082] In one embodiment, after constructing a nomogram for a target factor, the method further includes: plotting a calibration curve, an ROC curve, and a decision curve of the nomogram to comprehensively evaluate the risk prediction model.

[0083] A nomogram is constructed using the results of multivariate analysis as a graphical representation that describes the relationship between the frequency of observations and the probability of detection. In this embodiment, a calibration curve is used to verify the calibration of the nomogram. In addition, the receiver operating characteristic curve (ROC) is used to evaluate the accuracy of discrimination, while the classification performance is evaluated by calculating the area under the curve (AUC). The clinical utility of the nomogram is evaluated by decision curve analysis (DCA). All statistical tests are two-tailed tests, and a P-value < 0.05 is considered statistically significant.

[0084] As Figure 3 shown, the calibration curve of the nomogram is presented. The ideal line represents ideal prediction, the apparent line represents the prediction curve, and the bias-corrected line represents the calibration curve. The bias-corrected line is close to the ideal line, indicating that the nomogram is well-calibrated. The figure shows that there is a good consistency between the predictions generated by the nomogram and the observed results. The corrected c-index for the discriminability of the nomogram in the figure is 0.747, which is an accurate and robust performance estimate. In addition, the calibration plot shows that the nomogram predictions fit well with the actual observed results of the Hosmer-Lemeshow test (χ 2 = 3.3909, P = 0.9075).

[0085] As Figure 4 shown, the ROC curve of the nomogram further validates the performance of the model. It is worth noting that the area under the curve (AUC) of the nomogram exceeds that of a single indicator (CVAI), indicating that the constructed nomogram is a reliable scoring system. The AUC of the constructed nomogram is 0.747 (95% CI: 78–82.4%).

[0086] As Figure 5 shown, decision curve analysis (DCA) is used to evaluate the clinical applicability of the diagnostic nomogram. The x-axis represents the risk threshold probability varying from 0 to 1, and the y-axis represents the net benefit calculated for a given threshold probability. "ALL" represents the net benefit of intervening in all patients, and "None" represents the net benefit of patients without intervention, whose net benefit is zero. The net benefit of DCA refers to the net effect after considering the benefits and losses brought by false positive and false negative results. The figure shows that if the threshold probability is greater than 0.1, using this nomogram to diagnose type 2 diabetes patients will provide more benefits than the "ALL" or "None" options; if the threshold probability is 0.46, type 2 diabetes patients will benefit from using this nomogram.

[0087] The present invention analyzes and evaluates the influencing factors of type 2 diabetes mellitus complicated with MCI, and for the first time reveals the correlation between CVAI and cognitive function in patients with type 2 diabetes mellitus. This finding indicates that CVAI is associated with impaired cognitive function of memory and frontal executive function. Further analysis shows that even after adjusting for age, gender, education level, diabetes duration, and hypertension, CVAI in patients with type 2 diabetes mellitus is significantly negatively correlated with MoCA and MMSE. In addition, through univariate and multivariate logistic regression analysis, it is found that CVAI is an independent risk factor for MCI in patients with type 2 diabetes. Based on the above findings, a nomogram model is used to quantitatively evaluate this risk, aiming to provide empirical evidence for the prevention and treatment of patients with diabetes-related cognitive impairment. Finally, the ROC curve and internal validation calibration curve show that this method exhibits satisfactory performance, and the DCA curve analysis shows that the nomogram has greater net benefit, reflecting its significant clinical value.

[0088] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are displayed in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same moment, but can be executed at different moments. The execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turn with at least a part of the steps or stages in other steps or other steps.

[0089] Based on the same inventive concept, the embodiments of the present application also provide a risk prediction system for type 2 diabetes mellitus complicated with mild cognitive impairment for implementing the above-mentioned risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment. The implementation solutions provided by this system for solving problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the risk prediction system for type 2 diabetes mellitus complicated with mild cognitive impairment provided below can refer to the limitations for the risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment in the above text, and will not be repeated here.

[0090] In one embodiment, a risk prediction system for type 2 diabetes mellitus complicated with mild cognitive impairment is provided, including: an input module for acquiring the collected data of target factors; a processing module for predicting the quantitative risk of cognitive impairment by combining the collected data with a nomogram; and an output module for outputting the quantitative risk assessment result.

[0091] By displaying the quantified risks output by the output module, the visualization of cognitive impairment is realized, which helps type 2 diabetes patients to timely detect potential risks, intervene in a timely manner, and delay the progression of the disease.

[0092] Among them, the data collection of the input module is based on a Web platform. Users need to manually fill in relevant information in the input module, including age, gender, education level, diabetes duration, hypertension status, and CVAI. These variables correspond to the respective parameters in the nomogram. The system will locate and calculate in the nomogram based on these data, and finally generate a risk assessment result of mild cognitive impairment (MCI), and provide a 95% confidence interval (CI). The whole process not only ensures the accuracy of the data, but also ensures the convenience of operation and the security of the data.

[0093] In one embodiment, the system includes an ethnic recognition module; the processing module is further configured to match a corresponding nomogram according to the ethnicity recognized by the ethnic recognition module, and predict the quantified risk based on the matched nomogram.

[0094] The ethnic recognition module can collect face images through a camera, and combine machine learning (such as the SE-ResNet network) to recognize the ethnicity, and automatically input the recognition result into the processing module. The face recognition module can also be obtained by accessing the hospital database and using genetic means for recognition, and also automatically input the recognition result into the processing module. The recognition result of the ethnic recognition module can also be directly returned to the input module to provide the function of manual confirmation and modification, and then the input module inputs the confirmed collected data into the processing module for further processing.

[0095] Each module in the above risk prediction system for type 2 diabetes combined with mild cognitive impairment can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so as to facilitate the processor to call and execute the operations corresponding to the above modules.

[0096] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in all the above method embodiments are implemented.

[0097] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in all the above method embodiments are implemented.

[0098] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in all the above method embodiments are implemented.

[0099] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions.

[0100] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0101] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0102] The embodiments described above merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A risk prediction method for type 2 diabetes mellitus complicated with mild cognitive impairment, characterized in that, The method includes: Obtaining data related to potential risk factors of subjects with type 2 diabetes, and classifying the subjects into an MCI group and an NC group through cognitive assessment; the potential risk factors include demographic characteristics and clinical characteristics; Screening target factors from the potential risk factors according to the data related to the potential risk factors and the results of cognitive assessment; Constructing a nomogram for the target factors, and predicting the quantitative risk of cognitive impairment according to the nomogram.

2. The method according to claim 1, wherein: Introducing race-specific factors and constructing the nomogram that matches different races.

3. The method according to claim 1, wherein The classifying the subjects into an MCI group and an NC group through cognitive assessment includes: Performing cognitive assessment by using one or more of the Montreal Cognitive Assessment, Mini-Mental State Examination, Clock Drawing Test, DST, VFT, TMTA, TMTB, AVLT-IR, AVLT-DR, LMT to obtain quantitative cognitive assessment results; Comparing the quantitative cognitive assessment results with a threshold, and classifying the subjects into an MCI group and an NC group according to the comparison results.

4. The method according to claim 1, characterized in that, The screening the target factors from the potential risk factors according to the data related to the potential risk factors and the results of cognitive assessment includes: Performing between-group difference analysis by using t-test, Mann-Whitney test or chi-square test to determine the differences corresponding to each of the potential risk factors between the MCI group and the NC group; Performing non-parametric Spearman correlation analysis to determine the correlation between each of the potential risk factors and cognitive function; Calculating VAI or CVAI according to the data related to the potential risk factors, and evaluating the relationship between VAI or CVAI and cognitive function by using a multivariable linear regression method; Screening the target factors by using univariate logistic regression and backward stepwise logistic regression analysis.

5. The method according to claim 1, characterized in that, The predicting the quantitative risk of mild cognitive impairment according to the nomogram includes: Obtaining the collected data of the research object regarding the target factors; Based on the collected data, determining the scores corresponding to each of the target factors by looking up the nomogram; Adding up the scores corresponding to each of the target factors to obtain a total score, and quantitatively calculating the risk of mild cognitive impairment according to the total score. The calculation formula is as follows: Risk=-7.92e-07×points 3 +0.000274248×points 2 -0.022018993×points+0.605121118; In the formula, points represents the total score, and Risk represents the quantitative risk of mild cognitive impairment.

6. A risk prediction system for type 2 diabetes mellitus complicated with mild cognitive impairment, based on the method according to any one of claims 1 to 5, characterized in that, The system includes: An input module for obtaining the collected data of the target factors; A processing module for predicting the quantitative risk of cognitive impairment according to the collected data in combination with the nomogram; An output module for outputting the quantitative risk.

7. The system according to claim 6, wherein: The system includes a race identification module; the processing module is further configured to match the corresponding nomogram according to the race identified by the race identification module, and predict the quantitative risk based on the matched nomogram.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images