Bayesian implementation of the Royston-Parmar survival model

Hello, I am a clinician. It would be nice for me to be able to select the degrees of freedom of the baseline hazard and covariate effects to allow for ver flexible curves (i.e., crossing survival curves). It would also be nice to have posterior probabilities of therapeutic effect, as well as survival differences, with their respective HDIs, to test posterior probabilities as a function of time.