| 摘要 |
A method of training a learning-capable system, such as a neural network, with particular application to the fields of prognosis, risk assessment, intervention/treatment selection, and treatment outcome prediction is disclosed. The method includes the steps of providing a predetermined training data set of input and outcome data for a predetermined number of subjects, augmenting the input and/or outcome data, and training the system using the augmented data. Augmenting includes the steps of estimating propensity score data for each subject depending on its input data, diving the propensity score data into at least two strata, assigning each subject to a stratum, and augmenting the input data of each subject by its propensity score data and/or its stratum data.
|
