Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article


Doctor of Philosophy


Medical Biophysics


St Lawrence, Keith


Although positron emission tomography (PET) remains the gold standard for quantifying cerebral blood flow (CBF) and cerebral metabolic rate of oxygen (CMRO2), the procedures are invasive and complex, requiring arterial sampling and accounting for blood-borne activity. Non-invasive methods generally require extracting an image-derived input function (IDIF) from the carotid arteries; however, these methods are sensitive to partial volume errors due to the poor spatial resolution of PET. Hybrid PET/magnetic resonance imaging (MRI) offers an alternative non-invasive approach by using MRI to calibrate PET data. PET/MR imaging of oxidative metabolism (PMROx) is a reference-based approach that uses whole-brain (WB) measurements of oxygen extraction fraction and CBF to calibrate [15O]O2-PET data. Similarly, PET/MR imaging of CBF (PMRFlow) requires only WB CBF to calibrate dynamic [15O]H2O-PET data. Both methods avoid the need for invasive arterial sampling while maintaining the ability to quantify CBF and CMRO2.

This thesis contains the theoretical framework of PMROx, along with error analyses and an assessment on PET data acquired from healthy participants (n = 10), followed by its validation in an animal model by comparison to PET-alone measurements (n = 8). Additionally, the MRI-perfusion technique arterial spin labeling was used to further simplify PMROx by replacing [15O]H2O-PET, and PMROx was demonstrated to be sensitive to anesthetics-induced changes in metabolism. The accuracy of a PMRFlow approach of obtaining the IDIF (PMRFlowIDIF) from the WB [15O]H2O time-activity curve (TAC) was assessed in a porcine model by comparison to PET-only CBF measurements (n = 12). The ability of the method to remove blood-borne signal was demonstrated by generating CBF images for healthy individuals (n = 13).

In summary, results demonstrated the feasibility of producing quantitative CBF and CMRO2 images by PET/MRI without the need for invasive blood sampling. PMROx circumvents many of the complexities of traditional PET CMRO2 imaging and has the potential to reduce PET imaging to [15O]O2 only. PMRFlowIDIF is an easy-to-implement method that benefits from the high signal-to-noise ratio of the WB TAC and does not require vessel segmentation. Future studies involving human participants are required to fully validate these approaches, both in the healthy brain and in disorders characterized by disruptions in cerebral hemodynamics and metabolism.

Summary for Lay Audience

The brain is a complex organ that requires constant supply of blood and oxygen. As a consequence, tissue damage can occur within minutes to hours when supply of oxygen is interrupted. Measurements of cerebral blood flow and oxygen consumption with medical imaging modalities play an important role in assessing diseases such as carotid artery stenosis, as well as in understanding a variety of disorders, such as diabetes, drug addiction, and cancer. Over the years, research has also focused on answering questions regarding its role in neurodegenerative diseases, such as Alzheimer’s and Parkinson’s disease.

Imaging blood flow and oxygen consumption in the human brain is challenging. The gold standard is an imaging modality known as PET; however, the procedure is complex, invasive, and not readily accessible. One alternative is an emerging imaging technology that incorporates PET and MRI into a single hybrid scanner that can simplify the procedure. The aim of this thesis is to present non-invasive and fast alternatives to the PET procedure by using MRI measurements as reference. Here, the theory behind the proposed approaches is presented, together with their validation in an animal model.

In summary, results demonstrate the possibility of imaging cerebral blood flow and oxygen use by PET/MRI. When used in research studies, these approaches have the potential to help us better understand the role of interruptions in energy delivery and production in cerebral disorders. They can also help researchers develop more accessible imaging tools.

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Included in

Biophysics Commons