1. Generation of synthetic CT images
Traditionally, radiotherapy planning is based on CT images because the gray values in a CT represent attenuation coefficients for x-rays. At the MR-Linac, the ideal workflow would be based on MR images only without the need to acquire an additional CT scan. To make that possible, we work on algorithms to generate artificial CT images from MR images. To that end we use artificial intelligence techniques such as generative adversarial networks. Watch this video for an introduction to our work in this field.
2. Adaptive fractionation
Being able to visualize changes in the patient geometry through daily MR imaging, in principle allows us to exploit interfraction motion of the tumor. Adaptive fractionation is one approach to that. Here, the idea is to deliver higher doses on days when the tumor is further away from the critical organ at risk, and decrease the dose on days when the geometry is unfavorable.
3. Quantification of changes during fractionated radiotherapy
Daily MR-imaging of the patients allows us to quantify changes in the tumor and the surrounding organs over the course of therapy, which was impractical without the MR-Linac. For example, we can quantify shrinkage of the tumor or surrounding organs. Ultimately, this may allows us to predict the the individual patient's response to radiotherapy, and possibly adapt the treatment.