PET imaging involves injection of a radiolabelled compound called a radioligand which binds to a target molecule of interest and measuring the emitted radioactivity over time. Pharmacokinetic compartmental models can be fit to these time series to derive estimates of the concentration of the target molecule for quantification. However these models are complex nonlinear models requiring the estimation of a relatively large number of parameters. To this end, we have developed a new approach called SiMBA, which is a Bayesian hierarchical model which is able to borrow strength between data from multiple individuals to effectively adaptively regularise estimates from these models. We are now looking to apply SiMBA to new radioligands which bind to molecules involved in neuroinflammation. However radioligands for these targets tend to exhibit more complex pharmacokinetics than traditional models are able to accommodate. However, using our SiMBA approach, these models become more identifiable. In this project, we will be working on developing, applying and evaluating new pharmacokinetic models estimated using SiMBA for these neuroinflammatory targets.