Motion uncertainty deblurring in motion corrected reconstruction for ?PET brain imaging of awake rats

The implementation and improvement of the spatial resolution of the reconstructions of a motion correction method for unconstrained and unanesthetized rats is presented for a small bore (16 cm diameter) Siemens Inveon microPET scanner. The motion scans are corrected using the LOR rebinning technique. After correction, the images still suffer from loss of spatial resolution compared to motion-free scans. We propose to improve the spatial resolution after motion correction by using a motion dependent and spatially variant kernel applied to the reconstruction using 3D deconvolution in the image space. The resolution kernel is calculated using the motion data. The method is applied in a microDerenzo phantom and an in vivo experiment. In the phantom experiment the resolution loss, compared to the motion-free scan that was observed in the motion corrected scan, was reduced by the proposed additional deconvolution. Moreover, the final image resolution was independent of the motion when using the motion dependent deconvolution..