Norway's second PET / MRI scanner was installed and put into use in the PET center at UNN in the spring of 2018. In the autumn of 2018, the PET center started examining prostate cancer patients using a new radioactive tracer PSMA (Prostate Specific Membrane Antigen) that binds specifically to cancer cells emanating from the prostate gland. The figure below shows a cross-section of the pelvis, and the prostate in particular, for one of the first PET examinations performed at the PET center in Tromsø with this tracer.
In addition to the PET / MR image itself, there is significant information in the PET / MR examination that is not visualized in the form of an image. This can be, for example, dynamic kinetics in tracer recordings from PET and morphological and functional information from different MRI sequences. The purpose of this project is to develop methods in artificial intelligence and machine learning where all information is used to uncover the extent and aggressiveness of prostate cancer in a precise and effective way. In the long run, machine learning in PET / MRI examinations for prostate cancer could provide sufficient information to replace the current invasive biopsy for the classification of aggression.
This project is a collaboration between the PET center, SPKI, the Machine Learning Group and Visual Intelligence at UiT. David Wrangborg, who is a physicist, is leading the project.
