Key words: data, Rapid Learning, cancer therapy

Immensely powerful knowledge is hidden in the vast amounts of information we produce daily. Not only from our public’s socio-economic behaviour and health status, which we increasingly share online, but also from the treatment data we generate and collect in our clinical routine. What if we could harvest and exploit this wealth to enable the right treatment for the right patient?

This thesis discusses several Rapid Learning components that are needed to provide decision support systems, which help to determine the optimal therapy for cancer patients, taking cost-effectiveness into account. By incorporating cost-effectiveness analyses, even proton therapy is an option.