Visualizing tumor environment with perfusion and diffusion MRI: Computational challenges

Visualizing tumor environment is a critical task for assessing treatment response as well as tailoring therapy to the individual by better understanding the viable, necrotic and hypoxic areas. While a number of imaging modalities can provide complementary information about the tumor composition, there are several constraints regarding radiation, cost and patient tolerance that dictate the need of non-invasive and cost-effective methods to be used for tumor imaging in the context of personalized medicine. In this paper we present some of the major challenges in imaging tumor environment using perfusion and diffusion Magnetic Resonance Imaging (MRI) based on the actual computational workflows and discuss important computational issues that affect the robustness, reproducibility as well as the clinical significance of the extracted clinical biomarkers.