Positron emission tomography (PET) is a medical imaging technique that allows researchers and clinicians to study how the brain functions in real time. By injecting a small amount of a radioactive tracer, PET can measure processes such as brain metabolism, neurotransmitter activity, and protein accumulation linked to diseases like Alzheimer’s disease.

To obtain accurate quantitative measurements from PET scans, it is usually necessary to measure how much tracer is present in the blood over time. Traditionally, this requires placing a catheter in an artery, an invasive and uncomfortable procedure that makes quantitative PET difficult to use routinely in clinical practice.

Recent advances in PET technology, particularly the development of long axial field-of-view (LAFOV) PET scanners, may help overcome this limitation. These scanners can image large parts of the body simultaneously, including both the brain and major blood vessels such as the aorta. This creates the opportunity to estimate tracer concentration directly from the images themselves, reducing or even eliminating the need for arterial blood sampling.

This thesis of Laura Lopes Gonçalves da Providência e Costa investigates how LAFOV PET scanners can support non-invasive brain PET quantification. It explores new methods for analysing brain PET data using image-based and minimally invasive approaches, with a focus on maintaining accuracy while simplifying scanning procedures. The work also evaluates how these scanners can enable lower radiation doses and improve the practicality of quantitative brain PET imaging.

The overall goal of this research is to develop quantitative PET methods that are easier to implement in clinical routine, reduce patient burden, and make advanced brain imaging more accessible for both research and healthcare applications.