Artificial Neural Networks (ANNs), from their conception in the 1940s to their resurgence in the 1980s, brought new perspectives to the modeling of complex problems. Standing out for their ability to capture non-linear patterns, artificial neural networks, when added to physical knowledge, can enhance the understanding of the behavior of medications in the human body. This includes the behavior of iodinated contrast media, aiming to improve diagnostic accuracy in medical examinations. In this work we investigated the ability of neural networks informed by physics to model, from experimental data, the pharmacokinetics of iodinated contrast media. To this end, three different differential models were analyzed and the parameters of each model were estimated from experimental data. The results found show the feasibility of using PINNs in modeling the pharmacokinetics of contrast media in the human body. Such results can significantly contribute to the optimization of contrast dose in medical examinations, ensuring adequate image enhancement and an improvement in diagnostic accuracy. It is noteworthy that the study demonstrated the effectiveness of PINNs in modeling and predicting pharmacokinetic behaviors of iodinated contrast agent, providing a promising solution for the development of safe and effective pharmaceutical formulations. The continued importance of applying PINNs in the healthcare sector is also highlighted, aiming for significant advances in clinical practice and patient treatment. However, more research is still needed to validate and improve these models, ensuring their reliability and clinical applicability.