In this paper we introduce a methodology for adapting a population model to the actual patient dynamics, i.e. an individualization of target controlled infusion (TCI) based infusion. The solution proposed in this paper is to use the effect site concentration (Ce) of the drug into the patient as a feedback signal and to adapt the parameters of the Hill curve (relating the BIS and Ce) during the induction phase, resulting in a patient-individualized closed loop control of anesthesia. This allows moving from conventional generic patient models for drug infusion regulatory loops to personalized medicine.