Research On Construction Of Biological Age Prediction Model For Elderly In Ningxia

Objective Biological age can determine the true aging status of individuals more effectively and accurate prediction of BA can help to formulate targeted preventive measures for older individuals at an early stage,further explore the relationship between biological age and biomedical indicators,and help to take further measures to slow down the aging rate of the elderly.Methods A total of 2113 health elders aged from January 1 to December 30 of 2014 in CLHLS were included as the modeling population,fasting venous blood and urine were collected to detect biomedical indicators,biological age of the elderly was calculated using multiple linear regression,principal component analysis,random forest and extreme gradient enhancement algorithms,respective biological model was established.The model data were validated and applied to a total of 862 elderly people aged 60 years or older in the health checkup population of Ningxia Xi Xia District from January 1 to December 30 in 2022,as the model application population to calculate biological age and identify relevant biomedical indicators in subgroups for analysis Results,The study used mean absolute error(MAE),root mean square error(RMSE),and correlation coefficient(R~2)to assess the efficacy of the model.Results The calculation method that ultimately determines the best biological age is the random forest method.Significant differences were found in the importance of biomedical indicators for each age group in different subgroups of elderly individuals(low age,middle age,high age).Conclusion The use of the random forest method to calculate the biological age of the elderly and analyze the relevant biomedical indicators can more accurately locate the high-risk groups in the elderly and provide some guidance on healthy aging which can be of great help in slowing down the aging of the elderly,the analysis of the importance of biomedical indicators differs in each subgroup and requires individualized analysis of the characteristics to further study the occurrence mechanism and the reasons for the differences.