Spatiotemporal Power of Positron Emission Tomography: Pushing the Limits of Poisson Statistics in High-Resolution Human Neurotransmission Studies

Abstract

Brain disorders involving dysfunctions in neurotransmissionconstitute one of the most prevalent health problems. Subtledisruptions in human neurotransmission can result in significantdysfunction of cognition, locomotion, or practically any facet ofhuman behaviour. In turn, homeostasis of a specific neurotransmitter system can often be retrieved through pharmacologicalor lifestyle interventions. At present, human neurotransmissioncan be best assayed using positron emission tomography (PET). To date, neurotransmitter-PET (nt-PET) has been employedto investigate neuroreceptor level phenomenon in human behavior/cognition as well as in treatment development. In thecurrent work the goal was to explore and enhance the temporalcapabilities of nt-PET, to allow better characterization of thetemporal facets of neurotransmission.<br/><br/>Main obstacles limiting temporal characterization stem fromthe poor signal-to-noise-ratio of the PET measurement. Inparticular, the limitations related to image reconstruction algorithms and in turn the benefits obtained through regionalanalysis were in the focus of the investigations in this work. Themain finding was that the best temporal resolution achieved using a commonly recommended iterative reconstruction methodwas insufficient for temporal characterization, while a newlydeveloped algorithm allowing analytical reconstruction showedbetter temporal resolution without decreasing signal-to-noiseratio. Furthermore, a novel atlas-based regional analysis methodwas found superior to the currently employed manual region-ofinterest definition.<br/><br/>The findings made through this work will directly assist theplanning of future neurotransmission studies, and it is wishedthat the observations in this work would spark new, more widespread interest on the application of nt-PET in e.g. cognitivestimulation studies.