Protein markers for predicting probability of developing metabolic syndrome in a future time period and uses thereof
By detecting 11 protein biomarkers and age characteristics in serum samples, a machine learning model was constructed to predict the probability of developing metabolic syndrome in the future. This solves the problem that existing technologies cannot predict the risk of disease in advance, and enables accurate prediction of future diseases and early intervention.
CN117310171BActive Publication Date: 2026-05-29WESTLAKE UNIV
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WESTLAKE UNIV
- Filing Date
- 2022-08-04
- Publication Date
- 2026-05-29
Smart Images
Figure CN117310171B_ABST
Abstract
The present application provides a kind of protein marker for predicting the probability of suffering from metabolic syndrome in future period and its application, 11 kinds of protein markers with representative meaning are screened out by combining recursive feature elimination algorithm based on shapley value and Bourta feature selection method based on shapley value;In addition, the age characteristics and gender characteristics of the subject are input into the LGBM model, the age characteristics are selected by using the RFE based on feature importance, and the relative expression amount of any combination of age characteristics and 11 kinds of protein markers is used to predict the probability of suffering from metabolic syndrome in future period of the subject, if the probability is higher, it means that the subject is more likely to suffer from metabolic syndrome in future period, at this time, early intervention strategy can be made for the subject, and the effect of prevention is achieved.
Need to check novelty before this filing date? Find Prior Art