System for predicting sarcopenia of Alzheimer's disease patient

Abstract

The invention discloses a system for predicting sarcopenia of an Alzheimer's disease patient, and relates to the technical field of medical diagnosis, and the system comprises a risk factor collection module which is used for collecting independent risk factors of the Alzheimer's disease patient, the independent risk factors comprise the elderly, serum albumin, the ratio of creatinine to cystatin C and a clinical decline scale score; the sarcopenia risk prediction module is connected with the risk factor acquisition module, a sarcopenia combined risk prediction model is built in the sarcopenia risk prediction module in combination with independent risk factors, and the sarcopenia combined risk prediction model utilizes the acquired independent risk factors of the Alzheimer's disease patient to predict the risk of the Alzheimer's disease. And predicting and outputting a risk probability value of sarcopenia of the patient suffering from the Alzheimer's disease. By using the prediction model, a clinician can more objectively judge whether an AD patient has sarcopenia or not and which risk factors specifically exist, so that individualized treatment can be better performed on the patient.

Description

Technical Field

[0001] The present invention relates to the field of medical diagnosis technology, and more particularly to a system for predicting sarcopenia in patients with Alzheimer's disease. Background Art

[0002] Sarcopenia is an age-related disease of muscle mass loss, accompanied by a decrease in muscle strength and / or physical function. Sarcopenia is closely related to a series of adverse clinical events, such as falls, functional decline, frailty and death. Therefore, sarcopenia is considered one of the most serious health threats to the elderly.

[0003] The diagnostic process of sarcopenia is relatively complicated. According to the diagnostic criteria of the Asian Working Group on Sarcopenia (AWGS) in 2019, the diagnosis of sarcopenia needs to combine three aspects: muscle mass, muscle strength and physical function performance. First, the Simple Five Item Scoring Scale for Sarcopenia (SARC-F) or the calf circumference of the high-risk group for sarcopenia were used to screen out "possible sarcopenia" patients. Then, multiple objective indicators such as the Short Physical Performance Battery (SPPB), hand grip strength measurement, 5 sitting time, 4-meter or 6-meter walking speed, etc. were used to confirm that these patients had decreased muscle strength. Finally, specific tests such as dual-energy X-ray absorptiometry (DXA) or bioelectrical impedance analysis (BIA) were used to determine whether the patient's skeletal muscle mass index (SMI) was less than the diagnostic standard value of sarcopenia (male: <7.0 kg / m 2 ; Female: <5.7kg / m 2 ) can finally confirm sarcopenia. If these confirmed sarcopenia patients also have poor physical function, they will be diagnosed with "severe sarcopenia". In addition, even if the patient is finally diagnosed with sarcopenia, given its multifactorial pathogenic mechanism, how to achieve individualized intervention and treatment is also a major problem. Shortcomings of existing diagnostic solutions: (1) The diagnostic process of sarcopenia is relatively complicated, and coupled with the poor cooperation of Alzheimer's disease (AD) patients, it is relatively difficult to implement, which is not conducive to outpatient AD patient screening; in addition, BIA and DXA specificity are difficult to be fully popularized and promoted in lower-level medical institutions. Summary of the invention

[0004] In order to overcome the defects existing in the above-mentioned prior art, the present invention discloses a system for predicting sarcopenia in patients with Alzheimer's disease. The present invention combines the clinical variables and serum biological markers of AD patients with sarcopenia to screen out independent risk factors for sarcopenia in AD patients, and further constructs a risk prediction model for sarcopenia in AD patients based on these independent risk factors, providing a simple and objective quantitative reference tool for early identification and intervention of sarcopenia in the AD population.

[0005] In order to achieve the above objectives, the technical solution adopted by the present invention is: A system for predicting sarcopenia in patients with Alzheimer's disease, comprising: A risk factor collection module is used to collect independent risk factors of Alzheimer's disease patients, including: advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score; The sarcopenia risk prediction module is connected to the risk factor collection module, and a sarcopenia joint risk prediction model is constructed by combining independent risk factors. The sarcopenia joint risk prediction model uses the collected independent risk factors of Alzheimer's patients to predict the risk probability value of sarcopenia in Alzheimer's patients and output it.

[0006] 1. Risk Factor Collection Module In the present invention, the risk factor collection module collects four independent risk factors, namely, advanced age, serum albumin, creatinine to cystatin C ratio, and clinical frailty scale score. The above independent risk factors can be collected using an electronic medical record system; the electronic medical record system is then connected to a sarcopenia joint risk prediction model established in a computer. The independent risk factors are input into the sarcopenia joint risk prediction model, and the sarcopenia joint risk prediction model in the computer automatically calculates and outputs the risk probability value of sarcopenia in patients with Alzheimer's disease. The system of the present invention is executed using a computer or other device, so the system of the present invention can be applied in industry.

[0007] 2. Sarcopenia Risk Prediction Module Preferably, in the sarcopenia combined risk prediction model of the sarcopenia risk prediction module, a nomogram is used to predict and output the risk probability value of sarcopenia in Alzheimer's disease patients based on the combined risk factors of advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score.

[0008] Preferably, the nomogram includes parallel individual score segments, multiple prediction index segments, total score segments and risk probability segments; The multiple prediction index segments include an age segment, a clinical frailty scale score segment, a serum albumin segment, and a creatinine to cystatin C ratio segment; scales are evenly set in the single score segment, the multiple prediction index segments, and the total score segment; The specific scale within the unidirectional prediction indicator segment corresponds to the specific unidirectional score of the single item score segment. The unidirectional scores corresponding to multiple prediction indicators are added together to obtain the total item score, and the total item score is mapped to the total item score segment. The risk probability value corresponding to the total item score on the risk probability segment is the risk probability value of sarcopenia in patients with Alzheimer's disease.

[0009] Preferably, in the nomogram: The single score line segment represents the single score corresponding to each prediction indicator under different value conditions, with a value range of 0-100; The age line segment represents the patient's age, with a value range of 30-100, 30 is aligned with 0 of the single-item score line segment, and 100 is aligned with 100 of the single-item score line segment; The clinical frailty scale score line segment represents the patient's clinical frailty scale score, with a value range of 0-5, 0 is aligned with the 0 of the single score line segment, and 5 is aligned with 65 of the single score line segment; The serum albumin line segment represents the patient's serum albumin content, with a value range of 70-20, 70 is aligned with 0 of the single-item score line segment, and 20 is aligned with 87.5 of the single-item score line segment; The creatinine to cystatin C ratio line segment represents the patient's creatinine to cystatin C ratio, with a value range of 130-0, 130 is aligned with 0 of the single score line segment, and 0 is aligned with 68 of the single score line segment; The total score line segment represents the total score obtained by adding the individual scores of the four predictive indicators of advanced age, serum albumin, creatinine to cystatin C ratio, and clinical frailty scale score, with a value range of 0-260; The risk probability line segment represents the risk probability value of sarcopenia in patients with Alzheimer's disease, which is ultimately obtained based on the four predictive indicators of advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score. The value range is 0.05-0.95, with 0.05 aligned with the total score line segment of 100 and 0.95 aligned with the total score line segment of 204.

[0010] Preferably, it also includes: A basic data collection module is used to collect basic data of Alzheimer's disease patients, wherein the basic data includes clinical data, laboratory index data and muscle data; a data classification and comparison module, connected to the basic data acquisition module, for dividing Alzheimer's disease patients into an Alzheimer's disease patient combined with sarcopenia group and an Alzheimer's disease patient without sarcopenia group according to muscle data, and comparing the clinical data and laboratory index data of the Alzheimer's disease patient combined with sarcopenia group and the Alzheimer's disease patient without sarcopenia group to obtain difference indicators; The data analysis module is connected to the data classification and comparison module, the basic data acquisition module, the risk factor acquisition module and the sarcopenia risk prediction module, and is used to analyze the clinical data and laboratory indicator data of the Alzheimer's disease patients with sarcopenia and the non-sarcopenia group, obtain independent risk factors for sarcopenia in Alzheimer's disease patients, and send the independent risk factors to the risk factor acquisition module and the sarcopenia risk prediction module.

[0011] 3. Basic Data Acquisition Module In the present invention, the basic data collection module is used to collect basic data of Alzheimer's patients, including clinical data, laboratory index data, muscle data, etc.

[0012] Preferably, the clinical data collected by the basic data collection module include patient age, clinical frailty scale score, cognition, and quality of life.

[0013] Preferably, the laboratory index data collected by the basic data collection module include vitamin levels, hormone levels, nutritional levels, metabolic levels, and inflammation levels. The nutritional levels include serum albumin, and the metabolic levels include the ratio of creatinine to cystatin C.

[0014] Preferably, the muscle data collected by the basic data collection module includes muscle strength, muscle mass, body function status, and limb muscle content.

[0015] Preferably, in the basic data acquisition module, the AWGS 2019 diagnostic criteria are used to evaluate the muscle strength, muscle mass, and physical function status of Alzheimer's patients; and the bioelectrical impedance analysis method is used to measure the limb muscle content of Alzheimer's patients.

[0016] 4. Data Classification and Comparison Module In the present invention, the data classification and comparison module is used to divide Alzheimer's disease patients into an Alzheimer's disease patient complex sarcopenia group and an Alzheimer's disease patient non-sarcopenia group according to the muscle data collected by the basic data collection module, and compare the clinical data and laboratory index data of the Alzheimer's disease patient complex sarcopenia group and the Alzheimer's disease patient non-sarcopenia group to obtain difference indicators.

[0017] In the present invention, the Alzheimer's disease patients with sarcopenia group refers to Alzheimer's disease patients suffering from sarcopenia; the Alzheimer's disease patients without sarcopenia group refers to Alzheimer's disease patients not suffering from sarcopenia.

[0018] In the present invention, the difference index obtained by the data classification comparison module is used to preliminarily screen the risk factors for sarcopenia in patients with Alzheimer's disease. The univariate regression analysis of the data analysis module obtains the risk factors again. Subsequently, the data analysis module performs a multivariate regression analysis on these difference indexes, taking into account multiple variables and controlling potential confounding factors, so as to more comprehensively understand the impact of these risk factors on sarcopenia.

[0019] 5. Data Analysis Module Preferably, the analysis of the data analysis module includes: Univariate regression analysis was performed on the clinical data and laboratory index data of the Alzheimer's disease patients with and without sarcopenia to obtain differential variables; in the univariate regression analysis, the single factor in the clinical data and laboratory index data was used as the independent variable, and the presence or absence of sarcopenia was used as the dependent variable; Binary multivariate regression analysis was performed on the differential variables to obtain independent risk factors for sarcopenia in patients with Alzheimer's disease. In the binary multivariate regression analysis, multiple factors in the differential variables were used as independent variables, and the presence or absence of sarcopenia was used as the dependent variable.

[0020] 6. Model Evaluation Module Preferably, it also includes a model evaluation module, which is connected to the sarcopenia risk prediction module and is used to evaluate the sarcopenia combined risk prediction model.

[0021] Preferably, the evaluation of the model evaluation module includes: The receiver operating characteristic curve was used to evaluate the diagnostic value of the nomogram and single indicator in the sarcopenia joint risk prediction model, and to verify the higher diagnostic efficacy of the nomogram; The bias correction curve was used to evaluate the fitting relationship between the predicted value and the true value of the sarcopenia joint risk prediction model; The clinical decision curve was used to evaluate the clinical application value of the sarcopenia combined risk prediction model in diagnosing sarcopenia in patients with Alzheimer's disease.

[0022] 7. Treatment Recommendation Module Preferably, it also includes a treatment suggestion module, which is connected to the sarcopenia risk prediction module and generates sarcopenia treatment suggestions for the predicted sarcopenia in Alzheimer's patients.

[0023] Preferably, in the treatment recommendation module, when the risk probability value output by the nomogram in the sarcopenia combined risk prediction model is high risk, the treatment recommendation module generates sarcopenia treatment recommendations, including: supplementing albumin or creatine.

[0024] In the present invention, the sarcopenia joint risk prediction model integrates two clinical variables and two laboratory indicators to help clinicians quickly and accurately assess the risk of sarcopenia in patients with Alzheimer's disease. Once a patient is determined to be at high risk, the treatment recommendation module can implement more specific and personalized treatment intervention measures based on the risk indicators provided by the model, such as supplementing albumin or creatine, to provide advice to doctors, who can refer to the advice to reduce their workload, effectively treat sarcopenia in patients with Alzheimer's disease, and improve the overall prognosis of patients.

[0025] In the present invention, the risk factor collection module, the sarcopenia risk prediction module, the basic data collection module, the data classification and comparison module, the data analysis module, the model evaluation module, the treatment suggestion module, etc. can all be implemented by a computer or other device, so the system of the present invention can be applied in industry. For example, the risk factor collection module and the basic data collection module can use the electronic medical record system to collect information; the sarcopenia risk prediction module can use the prediction algorithm or model connected to the electronic medical record system for calculation (for example, including the nomogram data visualization tool); the data classification and comparison module can be processed by data mining tools or statistical software; the data analysis module can be analyzed by statistical analysis software or data analysis platform; the model evaluation module can be verified by evaluation tools or performance indicator calculation software; the treatment suggestion module can generate a treatment plan in combination with the intelligent decision support system and the medical expert system generated by the sarcopenia joint risk prediction model.

[0026] Beneficial effects of the present invention: 1. The combined presence of sarcopenia is significantly correlated with the poor clinical manifestations of AD patients. In view of its complex diagnostic process and multifactorial pathogenic mechanism, it is a difficult problem faced by neurologists to quickly identify and implement individualized intervention for sarcopenia in clinical work. However, there is currently no reference model available for clinicians to use. The present invention comprehensively investigates the common risk factors of AD and sarcopenia, including relevant laboratory indicators under vitamin levels, inflammation levels, nutritional metabolism levels, and hormone levels, and combines the clinical characteristics and scale evaluation data of AD patients to screen out 4 independent risk factors related to the occurrence of sarcopenia, and establishes a risk prediction model for sarcopenia in AD patients. It is finally verified that the prediction model has good predictive performance and clinical application value. By using this prediction model, clinicians can more objectively judge whether AD patients have sarcopenia and what specific risk factors exist, so as to better implement individualized treatment for patients. 2. The four predictive indicators finally included in this model are simple and easy to obtain. It only takes a few minutes to complete the screening of sarcopenia, which is convenient for promotion and use in medical institutions at all levels in the future.

[0027] 3. This model has passed internal validation, the model has high credibility and clear diagnostic value. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Schematic diagram of the system for predicting sarcopenia in patients with Alzheimer's disease according to the present invention.

[0029] Figure 2 This is a nomogram of the sarcopenia combined risk prediction model of the present invention.

[0030] Figure 3 Schematic diagram of the receiver operating characteristic curve (ROC) of the sarcopenia joint risk prediction model of the present invention and a single indicator.

[0031] Figure 4 This is a schematic diagram of the verification of the sarcopenia combined risk prediction model of the present invention.

[0032] Figure 5 It is a schematic diagram of the clinical decision curve analysis (Decision Curve Analysis, DCA) of the nomogram of the present invention. DETAILED DESCRIPTION

[0033] The concept, specific structure and technical effects of the present invention will be clearly and completely described below in conjunction with the embodiments and drawings to fully understand the purpose, characteristics and effects of the present invention.

[0034] A system for predicting sarcopenia in patients with Alzheimer's disease, e.g. Figure 1 As shown, including: A risk factor collection module is used to collect independent risk factors of Alzheimer's disease patients, including: advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score; a sarcopenia risk prediction module, connected to the risk factor collection module, wherein a sarcopenia joint risk prediction model is constructed by combining independent risk factors, and the sarcopenia joint risk prediction model predicts and outputs the risk probability value of sarcopenia in Alzheimer's patients by using the collected independent risk factors of Alzheimer's patients; A basic data collection module is used to collect basic data of Alzheimer's disease patients, wherein the basic data includes clinical data, laboratory index data and muscle data; a data classification and comparison module, connected to the basic data acquisition module, for dividing Alzheimer's disease patients into an Alzheimer's disease patient combined with sarcopenia group and an Alzheimer's disease patient without sarcopenia group according to muscle data, and comparing the clinical data and laboratory index data of the Alzheimer's disease patient combined with sarcopenia group and the Alzheimer's disease patient without sarcopenia group to obtain difference indicators; A data analysis module, connected to the data classification and comparison module, the basic data collection module, the risk factor collection module and the sarcopenia risk prediction module, for analyzing the clinical data and laboratory index data of the Alzheimer's disease patients with sarcopenia and the non-sarcopenia group, obtaining independent risk factors for sarcopenia in Alzheimer's disease patients, and sending the independent risk factors to the risk factor collection module and the sarcopenia risk prediction module; A model evaluation module, which is connected to the sarcopenia risk prediction module and is used to evaluate the sarcopenia combined risk prediction model; A treatment suggestion module is connected to the sarcopenia risk prediction module to generate sarcopenia treatment suggestions for the predicted sarcopenia in the Alzheimer's disease patient.

[0035] In this embodiment, the risk factor collection module is used to collect four independent risk factors: advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score.

[0036] For the basic data collection module, the researchers completed the clinical scale assessment of all AD patients in a face-to-face manner; collected laboratory indicators of all patients, including vitamin levels, hormone levels, nutritional levels, metabolic levels, and inflammatory levels; used the AWGS 2019 diagnostic criteria to evaluate muscle strength, muscle mass, and physical function of all included AD patients; and measured the limb muscle content of AD patients by the BIA method. The collected clinical data, laboratory indicator data, and muscle data were entered into the computer to complete the basic data collection work.

[0037] For the data classification comparison module, the clinical variables and laboratory indicators of patients in the AD combined with sarcopenia group (ADSA) and AD without sarcopenia group (ADNSA) were compared and difference indicators were obtained. It was also found that patients in the ADSA group had worse cognitive function, emotional state and quality of life than those in the ADNSA group, and cognitive impairment involved multiple cognitive domains. Even after adjusting for age and diabetes ratio, these differences still existed, confirming that the presence of sarcopenia is an independent factor for poor clinical manifestations in AD patients.

[0038] For the data analysis module, all clinical variables and laboratory indicators were subjected to univariate regression analysis to obtain differential variables, and binary multivariate regression analysis was further performed on the differential variables. The presence or absence of sarcopenia was used as the dependent variable, and finally it was found that advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score were independent risk factors for sarcopenia in AD patients. The four independent risk factor indicators obtained were sent to the risk factor collection module and the sarcopenia risk prediction module. The sarcopenia risk prediction module uses the four independent risk factor indicators to construct a sarcopenia joint risk prediction model; the risk factor collection module collects the specific values ​​of the four independent risk factor indicators, and the sarcopenia joint risk prediction model uses the specific values ​​of the four independent risk factor indicators to calculate the risk probability value of sarcopenia in Alzheimer's patients.

[0039] For the sarcopenia risk prediction module, a nomogram is used in the sarcopenia joint risk prediction model. The nomogram is based on the combined risk factors of advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score to predict and output the risk probability value of sarcopenia in patients with Alzheimer's disease.

[0040] Nomogram Figure 2 As shown in the figure, the line segment corresponding to each predictor variable is marked with a corresponding scale, which represents the range of values ​​of the predictor variable, and the length of the line segment reflects the contribution of the predictor variable to the occurrence of sarcopenia in AD patients. The "Points" (first row) in the figure represent the individual scores corresponding to each predictor variable under different values, "Total points" represents the total score obtained by adding the individual scores of all the above four variables (the second to last row), and "Risk of events" represents the risk probability corresponding to "Total points", that is, the risk probability value of sarcopenia in AD patients finally obtained based on the four predictive indicators. Age represents age; FRS represents clinical frailty scale score; Albumin represents serum albumin; CCR represents creatinine to cystatin C ratio.

[0041] like Figure 2 As shown in the Nomogram diagram: The single score line segment represents the single score corresponding to each prediction indicator under different value conditions, with a value range of 0-100; The age line segment represents the patient's age, with a value range of 30-100, 30 is aligned with 0 of the single-item score line segment, and 100 is aligned with 100 of the single-item score line segment; The clinical frailty scale score line segment represents the patient's clinical frailty scale score, with a value range of 0-5, 0 is aligned with the 0 of the single score line segment, and 5 is aligned with 65 of the single score line segment; The serum albumin line segment represents the patient's serum albumin content, with a value range of 70-20, 70 is aligned with 0 of the single-item score line segment, and 20 is aligned with 87.5 of the single-item score line segment; The creatinine to cystatin C ratio line segment represents the patient's creatinine to cystatin C ratio, with a value range of 130-0, 130 is aligned with 0 of the single score line segment, and 0 is aligned with 68 of the single score line segment; The total score line segment represents the total score obtained by adding the individual scores of the four predictive indicators of advanced age, serum albumin, creatinine to cystatin C ratio, and clinical frailty scale score, with a value range of 0-260; The risk probability line segment represents the risk probability value of sarcopenia in patients with Alzheimer's disease, which is ultimately obtained based on the four predictive indicators of advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score. The value range is 0.05-0.95, with 0.05 aligned with the total score line segment of 100 and 0.95 aligned with the total score line segment of 204.

[0042] For the model evaluation module: The receiver operating curve results showed that the prediction model established by combining these four risks had the greatest diagnostic value [area under the curve (AUC)] = 0.822, 0.822 (95% CI: 0.786-0.883, p <0.001)), with a sensitivity of 80.6% and a specificity of 75.5%. Figure 3 As shown. Among them, the AUC values ​​of the other single predictive factors from high to low are age at diagnosis [(AUC=0.730, 95% Confidence Interval (CI): 0.679-0.782, p <0.001)], creatinine to cystatin C ratio (AUC=0.687, 95% CI:0.625-0.750, p <0.001), Clinical Frailty Scale score (AUC=0.669, 95% CI: 0.613-0.724, p <0.001)and serum albumin (AUC=0.660, 95% CI: 0.597-0.723, p <0.001). The bootstrap test was repeated 1000 times for the model established by the combined indicators. The results showed that the combined prediction model had a good fit and good calibration, such as the bias correction curve Figure 4 The decision curve analysis (DCA) curve shows that the combined prediction model has practical clinical application value in diagnosing sarcopenia in AD patients, as shown in Figure 2. Figure 5 shown.

[0043] Figure 3 In the figure, Model represents the ROC curve of the nomogram for diagnosing sarcopenia, showing the ability to identify sarcopenia in patients with Alzheimer's disease; Age represents the ROC curve of age for diagnosing sarcopenia; CCR represents the ROC curve of creatinine to cystatin C ratio for diagnosing sarcopenia; FRS represents the ROC curve of clinical frailty scale score for diagnosing sarcopenia; Albumin represents the ROC curve of serum albumin level for diagnosing sarcopenia. In the figure, Specificity represents specificity, Sensitivity represents sensitivity, and AUC represents the area under the curve.

[0044] Figure 4 In the figure, the x-axis represents the probability predicted by the nomogram, and the y-axis represents the actual probability of sarcopenia in patients with Alzheimer's disease. The black dashed line represents the ideal case of perfect prediction; the green dashed line represents the prediction performance of the entire cohort (n=368); the red solid line represents the observed probability of sarcopenia after bootstrap correction (B = 1000 repetitions), which represents the observed performance of the nomogram.

[0045] Figure 5 In the figure, High Risk Threshold represents the risk probability threshold for sarcopenia in AD patients; Standardized Net Benefit represents the clinical net benefit of sarcopenia intervention; Model represents the DCA curve based on the sarcopenia combined risk prediction model; All represents the DCA curve that takes sarcopenia into account in all AD patients; None represents the DCA curve that does not take sarcopenia into account in AD patients.

[0046] Therefore, if Figure 3 As shown in Figure 2, the diagnostic value of this prediction model for diagnosing sarcopenia is the greatest. Figure 4 This indicates that the predicted values ​​of the model are well matched to the true values. Figure 5 It is suggested that when the threshold probability is 20%~100%, the prediction model shows a good positive net benefit, indicating that the prediction model has good potential clinical application value.

[0047] The key points of the present invention are as follows: 1) The present invention is based on studying the relevant risk factors of sarcopenia in a large sample of AD population, and constructing a prediction model in combination with serum biological markers. Through this prediction model, clinicians can more objectively judge whether AD patients have sarcopenia and what specific risk factors exist, so as to better implement individualized treatment for patients.

[0048] 2) This invention uses the clinical frailty rating scale (FRS) for the first time to quickly assess the frailty status of AD patients, and found that a lower FRS score is an independent risk factor for sarcopenia in AD patients, and has good value in predicting sarcopenia. In the future, more inventions are needed to explore the predictive value of the FRS scale for sarcopenia in online AD patients and its clinical application prospects.

[0049] 3) The present invention found that low albumin (ALB) levels are a risk factor for sarcopenia in AD patients. After adjusting for age and diabetes ratio at diagnosis, partial correlation analysis found that ALB levels were correlated with both sarcopenia assessment indicators and cognitive assessment indicators of patients. Our invention suggests that lower ALB levels can not only predict the occurrence of sarcopenia in AD, but also affect the cognitive function of AD patients, and has certain disease specificity. In the future, reasonable supplementation of ALB can not only reduce the occurrence of sarcopenia in AD patients, but also may delay the progression of AD.

[0050] The above is a specific description of the implementation mode of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention, and these equivalents or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A system for predicting sarcopenia in patients with Alzheimer's disease, characterized in that include: A risk factor collection module is used to collect independent risk factors of Alzheimer's disease patients, including: advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score; The sarcopenia risk prediction module is connected to the risk factor collection module, and a sarcopenia joint risk prediction model is constructed by combining independent risk factors. The sarcopenia joint risk prediction model uses the collected independent risk factors of Alzheimer's patients to predict the risk probability value of sarcopenia in Alzheimer's patients and output it.

2. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 1, characterized in that: In the sarcopenia combined risk prediction model of the sarcopenia risk prediction module, a nomogram is used to predict and output the risk probability value of sarcopenia in patients with Alzheimer's disease based on the combined risk factors of advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score.

3. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 2, characterized in that: The nomogram includes parallel individual score segments, multiple prediction index segments, total score segments and risk probability segments; The multiple prediction index segments include an age segment, a clinical frailty scale score segment, a serum albumin segment, and a creatinine to cystatin C ratio segment; scales are evenly set in the single score segment, the multiple prediction index segments, and the total score segment; The specific scale within the unidirectional prediction indicator segment corresponds to the specific unidirectional score of the single item score segment. The unidirectional scores corresponding to multiple prediction indicators are added together to obtain the total item score, and the total item score is mapped to the total item score segment. The risk probability value corresponding to the total item score on the risk probability segment is the risk probability value of sarcopenia in patients with Alzheimer's disease.

4. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 3, characterized in that: In the nomogram: The single score line segment represents the single score corresponding to each prediction indicator under different value conditions, with a value range of 0-100; The age line segment represents the patient's age, with a value range of 30-100, 30 is aligned with 0 of the single-item score line segment, and 100 is aligned with 100 of the single-item score line segment; The clinical frailty scale score line segment represents the patient's clinical frailty scale score, with a value range of 0-5, 0 is aligned with the 0 of the single score line segment, and 5 is aligned with 65 of the single score line segment; The serum albumin line segment represents the patient's serum albumin content, with a value range of 70-20, 70 is aligned with 0 of the single-item score line segment, and 20 is aligned with 87.5 of the single-item score line segment; The creatinine to cystatin C ratio line segment represents the patient's creatinine to cystatin C ratio, with a value range of 130-0, 130 is aligned with 0 of the single score line segment, and 0 is aligned with 68 of the single score line segment; The total score line segment represents the total score obtained by adding the individual scores of the four predictive indicators of advanced age, serum albumin, creatinine to cystatin C ratio, and clinical frailty scale score, with a value range of 0-260; The risk probability line segment represents the risk probability value of sarcopenia in patients with Alzheimer's disease, which is ultimately obtained based on the four predictive indicators of advanced age, serum albumin, creatinine to cystatin C ratio and clinical frailty scale score. The value range is 0.05-0.95, with 0.05 aligned with the total score line segment of 100 and 0.95 aligned with the total score line segment of 204.

5. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 1, characterized in that: Also includes: A basic data collection module is used to collect basic data of Alzheimer's disease patients, wherein the basic data includes clinical data, laboratory index data and muscle data; a data classification and comparison module, connected to the basic data acquisition module, for dividing Alzheimer's disease patients into an Alzheimer's disease patient combined with sarcopenia group and an Alzheimer's disease patient without sarcopenia group according to muscle data, and comparing the clinical data and laboratory index data of the Alzheimer's disease patient combined with sarcopenia group and the Alzheimer's disease patient without sarcopenia group to obtain difference indicators; The data analysis module is connected to the data classification and comparison module, the basic data acquisition module, the risk factor acquisition module and the sarcopenia risk prediction module, and is used to analyze the clinical data and laboratory indicator data of the Alzheimer's disease patients with sarcopenia and the non-sarcopenia group, obtain independent risk factors for sarcopenia in Alzheimer's disease patients, and send the independent risk factors to the risk factor acquisition module and the sarcopenia risk prediction module.

6. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 5, characterized in that: The clinical data collected by the basic data collection module include patient age, Clinical Frailty Scale score, cognition, and quality of life; The laboratory index data collected by the basic data collection module include vitamin level, hormone level, nutritional level, metabolic level, and inflammation level. The nutritional level includes serum albumin, and the metabolic level includes the ratio of creatinine to cystatin C. The muscle data collected by the basic data collection module include muscle strength, muscle mass, body function status, and limb muscle content; In the basic data acquisition module, the AWGS 2019 diagnostic criteria are used to evaluate the muscle strength, muscle mass, and physical function status of patients with Alzheimer's disease; and the bioelectrical impedance analysis method is used to measure the limb muscle content of patients with Alzheimer's disease.

7. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 5, characterized in that: The analysis of the data analysis module includes: Univariate regression analysis was performed on the clinical data and laboratory index data of the Alzheimer's disease patients with and without sarcopenia to obtain differential variables; in the univariate regression analysis, the single factor in the clinical data and laboratory index data was used as the independent variable, and the presence or absence of sarcopenia was used as the dependent variable; Binary multivariate regression analysis was performed on the differential variables to obtain independent risk factors for sarcopenia in patients with Alzheimer's disease. In the binary multivariate regression analysis, multiple factors in the differential variables were used as independent variables, and the presence or absence of sarcopenia was used as the dependent variable.

8. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 1, characterized in that: It also includes a model evaluation module, which is connected to the sarcopenia risk prediction module and is used to evaluate the sarcopenia combined risk prediction model.

9. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 8, characterized in that: The evaluation of the model evaluation module includes: The receiver operating characteristic curve was used to evaluate the diagnostic value of the nomogram and single indicator in the sarcopenia joint risk prediction model, and to verify the higher diagnostic efficacy of the nomogram; The bias correction curve was used to evaluate the fitting relationship between the predicted value and the true value of the sarcopenia joint risk prediction model; The clinical decision curve was used to evaluate the clinical application value of the sarcopenia combined risk prediction model in diagnosing sarcopenia in patients with Alzheimer's disease.

10. A system for predicting sarcopenia in patients with Alzheimer's disease as claimed in claim 1, characterized in that: It also includes a treatment suggestion module, which is connected to the sarcopenia risk prediction module and generates sarcopenia treatment suggestions for the predicted sarcopenia in Alzheimer's patients.

Images