Feature construction method, system, device and medium based on metabolic comprehensive index

By constructing a comprehensive metabolic index, the problem that machine learning technology cannot describe the development process of risk factors in the prediction of metabolic syndrome is solved, thereby improving the accuracy of prediction and the effectiveness of intervention.

CN115458163BActive Publication Date: 2026-06-23SOUTH CHINA NORMAL UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SOUTH CHINA NORMAL UNIV
Filing Date
2022-08-11
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

Existing machine learning techniques cannot effectively summarize and describe the development process of risk factors in predicting metabolic syndrome, nor can they understand the differences in disease risk caused by the annual changes of each indicator at different numerical levels.

Method used

By acquiring raw indicators, preprocessing them to construct structured data, calculating differential features and obtaining weighting functions, a comprehensive metabolic index is constructed to predict the comprehensive metabolic disease status in future time periods.

Benefits of technology

It enables a better summary and description of the development process of risk factors, improves the accuracy and targeting of metabolic syndrome prediction, and provides better intervention and treatment options.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115458163B_ABST
    Figure CN115458163B_ABST
Patent Text Reader

Abstract

The application discloses a feature construction method, system and device based on metabolic comprehensive indexes and a medium, and can be applied to the technical field of metabolic data processing. According to the method, after the obtained original indexes are preprocessed, a sample data set of structured data including continuous time periods is constructed according to the structured data obtained through the preprocessing, the differential features of the sample data set are calculated, a weight function used for evaluating the importance when different numerical values change is obtained, then the metabolic comprehensive indexes are constructed according to the weight function and the differential features, so that the metabolic comprehensive indexes know the different performances of the annual changes of each index at different numerical levels on the risk of diseases, and the development process of the risk factors is better summarized and described.
Need to check novelty before this filing date? Find Prior Art