"I'm doing research on almost all aspects of adaptive radiotherapy. A nice example is the MR-linac, a device that has a lot of potential. It integrates the irradiation device with an MRI scanner. So far, we have worked a lot with a device that combines a CT scanner with an irradiation device. With MRI, we can distinguish the tumour much better from the healthy tissue. The technical hypothesis is that, with this improved image quality and good algorithms, we can adjust the treatment plans on a daily basis. A few years ago, we started studying it to make sure that we would be up and running once the device was put into operation. In the course of the summer we will treat the first patients. This makes it possible to gather a lot of data to refine the algorithms and enable more advanced treatments. Exploring the possibilities of such a device I find enormously motivating. With functional MRI, you could also start looking at the biology of the tumour and how it changes during treatment. Furthermore, we have developed software that can make videos instead of static images. We would like to add that software to the MR-linac so that we can visualize movements and adjust the treatment accordingly."
"That is the use of medical imaging to guide therapy to the right place. Surgery and radiotherapy are two forms of therapy that are ideally suited for this. Using images made in the treatment room, you can ensure that you only irradiate the tumour and affect the healthy tissue as little as possible. So you have to hit the mark. Surgeons also need scans in order to be able to navigate properly to the tumour and to remove sufficient, but also not too much, tissue.
In addition, you can use imaging to measure whether a certain therapy is successful and whether adjustment is needed. Hopefully, with radiotherapy, we can increase or decrease the intensity of the radiation locally on the basis of images that show which part of the tumour responds well or less well."
I think that artificial intelligence will be an important driver for the successful further introduction of image-guided therapies, such as autonomous analysis and registration of scans, recording images and making decisions. You would ultimately like to continuously adapt a treatment plan to changes during treatment. That cannot be done fast enough at the moment. We need a system that can continually, and extremely accurately, monitor the anatomy of the patient and adjust the treatment by itself. What we need is a kind of route planner that can recalculate the route during a treatment if the situation changes."
Furthermore, I am convinced that artificial intelligence will have a major impact on medicine. For many doctors, for example, it was an eye-opener that an algorithm has been developed that could detect melanomas on the photos of the skin just as well as the most experienced dermatologists. Such developments offer many possibilities but will also cause a lot of uproar. Do we dare to trust such algorithms and are our jobs still secure?"