Recent developments on multifunctional nano-systems have opened new perspectives for tumor control by proposing new nano-actuators and nano-sensors in in vivo anti-cancer treatments. But the delivery control of these nano-agents into the cancer cells is one of the major factors that directly affect the efficiency of nanotherapies. In this study, we show that system identification methods (CONTSID Matlab toolbox), generally used in control engineering, can bring efficient solutions to help biologists to estimate crucial parameters of the nanoparticles pharmacokinetics from experimental data. The in vivo results presented herein clearly emphasize the relevance of these data-driven modeling approaches associated with magnetic resonance imaging.