Research

Below are the research projects of the Catana Lab. Please click on the figures for more details.

Attenuation Correction

Alt text Attenuation is caused by the interaction of the 511 keV photons with electrons in tissue, leading to inaccurate quantification. Challenges arise in using MR as opposed to CT for derivation of linear attenuation coefficients for PET reconstruction. In our lab we have developed methods combining strengths from atlas-based methods and tissue segmentation methods to create accurate and reproducible voxel-wise linear attenuation coefficient maps (ยต-maps) of the head.

Partial Volume Effects Correction

Alt text Partial volume effects (PVEs) arise due to the limited spatial resolution of the PET scanner and the discretization of the reconstructed image into voxels. This coupled with the dimensions of the object imaged will result in quantification discrepancies of the activity in an imaged subject. In our lab we have investigated the point-spread function of our BrainPET scanner. In addition, MR-assisted methods are implemented to correct for PVEs during image reconstruction and as a post-processing step; we are also investigating the improvement of image quality due to these methods.

Motion Correction

Alt text PET-based and MR-based methods

Image-derived Arterial Input Function

Alt text Determination of the arterial input function (AIF) is an important step in order to asses pharmacokinetic parameters from dynamic data using a kinetic model. Image-based AIF (I-AIF) estimation uses MR and PET images to non-invasively derive the AIF.

Optimized PET Data for Dementia Studies

Alt text Imaging dementia subjects is challenging for PET studies since the subjects are prone to motion and have severe cortical atrophy. In our lab we are applying MR-assisted PET corrections to optimize the simultaneously-acquired MR and PET data. The optimized data will then be evaluated for use as multimodal biomarkers in these patients.