Construction of the data-driven transformation risk prediction model for young and middle-aged prediabetic patients
HU Shengying
ZHANG Yizhu
SUN Yumei
SUN Hongyu
Abstract：Objective:To construct a transformation risk prediction model for young and middle-aged prediabetic patients based on the data from the health information platform of Yinzhou district of Ningbo city Zhejiang province.Methods:A retrospective cohort study design was adopted.Combined with literature review and expert recommendations,24 predictors were included.LASSO regression was applied for variable screening.The prediction model was constructed by using Logistic regression and four machine learning algorithms.Results:A total of 7 379 patients were included,among whom 731 cases(9.91%)transformed into type 2 diabetes.The random forest model performed relatively well.It had a good discriminatory ability.However,its accuracy was moderate.Conclusions:The transformation risk prediction model for young and middle-aged prediabetic patients based on random forests has good discriminatory ability.It could provide a reference for optimizing the precise management of young and middle-aged prediabetic patients.
Keywords：middle-aged and young peopleprediabetestype 2 diabetes mellitusmachine learningpredictive modelinfluencing factorsnursing
Publication Date：2025-12-25
Online Publishing Date：2025-12-30(First online date of this platform, not the publication date of the document)
Pages：8( 4087-4094 )