For evaluation indicators in prognostication

[itoi9350]

I have a question regarding the metrics in prognostication. The paper states that you used the median risk score, which is the output of the model, to separate high-risk patients from low-risk patients. This was done with the data set used in the experiment and a log-rank test was performed on the high-risk patient group and the low-risk patient group to determine if there was a significant difference. I understand this to be the case. Is it correct to assume that the Kaplan-Meier curves were drawn based on the information on life and death that was originally included in the data set as teacher data? Also, you used AUROC to compare the performance of the prognostic model with other basic models.

I would appreciate it if you could help me with this.

[Dadatata-JZ]

@itoi9350 Hi there,

Not quite sure what you meant by "teacher data". The KM curves were drawn using the survival data after the high-risk and low-risk groups were decided by the model using histology images. Lmk.