Imaging of pre- and post-treatment H&N cancer: Technical aspects, radio(geno)mics and application in chemoradiaton and immunotherapy

This thesis investigated the role of advanced magnetic resonance imaging (MRI) techniques in the diagnosis and treatment of head and neck cancer. The research was divided into three main parts: technical optimization, treatment prediction, and decoding tumor biology. First, the thesis demonstrated that a specific imaging technique (DW-SPLICE) provides clearer images with less distortion compared to standard methods, making it easier to detect if a tumor has returned. Additionally, it introduced an artificial intelligence tool that can automatically and accurately outline tumors on scans, a task that is currently time-consuming for doctors. The research also explored how MRI can predict which patients will benefit most from conventional treatments or newer options like immunotherapy. And tried to see if MRI texture analyses can give more insight into the early response to immunotherapy. Finally, the thesis showed that MRI can serve as a "virtual biopsy," non-invasively revealing biological characteristics of a tumor that usually require invasive tissue samples. Overall, this work contributes to more personalized and effective care for patients with head and neck cancer.