PET Methodology

Positron emission tomography (PET) is a tomographic technique that, through a ring of detectors, measures the fate of natural molecules labelled with positron emitters (18F, 11C, 15O). The high-energy of the emitting radiation allows the injection of minimal quantities of these compounds so that the organ or process measured is not perturbed. PET measures allow the formation of quantitative images of rates of physiological processes in vivo. In contrast to structural imaging (CT, MRI), PET can be used to examine dynamic changes in selected functional processes, and its quantitative potential not only enables the localisation of functional changes but also the assessment of their magnitude. The use of PET is becoming common in research, clinical diagnostic and drug development. Transformation of regional tissue radioactivity measured by the PET scanner into the rate of a physiological process requires use of sophisticated designs, accurate calibrations and computational models to describe the radiation detection process as well as the kinetics of the radiotracer in the organ of interest. The groups focus is on the development of methodology for the quantitative estimation of physiological parameters from PET studies. Current research focuses on computational chemistry, neuro-informatics, image reconstruction, kinetic modelling and multimodal image integration.

Background & Relevance|PET allows the formation of quantitative images of rates of physiological or biochemical processes in vivo. In contrast to structural imaging (CT, MRI), PET can be used to examine dynamic changes in selected functional processes, and its quantitative potential not only enables the localisation of functional changes but also the assessment of their magnitude. Currently in the UK there is an unprecedented investment in PET technology from academia and industry with two new major PET imaging centres recently opened (the GSK Clinical Imaging Centre at Hammersmith Hospital, London and the Wolfson Molecular Imaging Centre at the University of Manchester) and three more in planning (Bristol, Cardiff and Kings College in London). These will be recent additions to the already significant presence of PET technology in healthcare in the UK||Research Focus|The group focuses on the development of computational methodology (experimental design, mathematical/statistical modelling), theoretical and applied, for the quantitative estimation of physiological parameters from PET studies. Current research focuses on:||Computational Chemistry (Dr. Lula Rosso)|One of the most common reasons for a new PET radiotracer candidate failure is that high non-specific and sub cellular binding obscures the specific binding to the molecular target and, thus, reduces the quality of the PET scan data. With the advance of high speed computer and new algorithms, it is now possible to investigate important processes of biological systems at microscopic level over increasingly longer times. The goal is to reproduce in silico non-specific binding trends of in vivo PET experiments investigating the mechanisms and dynamics of lipid-radiotracer interaction. Eventually, this could provide a very useful screening tool to reduce potential ligands failure rate after in vivo PET evaluation.||Neuroinformatics (Dr. Subrata Bose)|The exquisite quantitative potential of PET makes this tomographic technique sensitive to the subtle changes in neuronal function that characterize psychiatric disorders. By combining the functional patterns detected by PET with ancillary data (family history, genetic profiling, proteomic and/or genomic data obtained from peripheral measures etc.) we aim to construct robust biomarkers of disease for diagnostic purposes. At the same time, PET can be used to validate peripheral biomarkers that can be used at a lower cost. Research focuses of the optimization of data-mining and multivariate statistical analysis techniques for application of pattern discovery in PET data.||Image Reconstruction/Corrections and Kinetic Modelling (Dr. Federico Turkheimer)|Transformation of regional tissue radioactivity measured by the PET scanner into the rate of a physiological or biochemical process requires use of sophisticated designs, accurate calibrations and computational models to describe the radiation detection process as well as the kinetics of the radiotracer in the organ of interest. The Unit is producing mathematical models and computational techniques of increasing complexity that account for the specific properties of scanners (geometry and sensitivity), the radiation emission process (scatter and random coincidences), the time activity of the tracer in relation to the physiology of the system (kinetic modelling), the nature of the organ imaged (multi-scale spatial modelling) and its anatomy (multimodal integration with CT and MRI images).