A disease risk determination method that combines survival analysis with downsampling of class imbalanced sets

摘要

A method for downsampling a class disparity set by survival analysis, comprising: obtaining a class disparity data set, the class disparity data set comprising biological data from a plurality of subjects, the biological data of each subject comprising observation, time values and a plurality of clinical measures, wherein the biological data is classified as being of a majority data class or part of a minority data class, wherein the majority data class has a greater number of observations than the minority data class; downsampling the class imbalanced data set, wherein the downsampling causes the majority data class to have an equal or substantially equal number of observations to the minority data class; and performing cross-validation on the data set downsampled by survival analysis to generate a survival model; Observations may or may not contain events at specific time values.