A software tool for merging two PET/CT body segments into a single scan

Learning Objectives Whole-body (WB) scans from head to toe are not feasible on certain PET/CT scanners and are divided into upper & lower body segments that are difficult to view as a single study. The aim of this abstract is to develop an automated software tool to stitch the two body segments and facilitate their interpretation.

Summary: 6 WB PET/CT scans were acquired on GE DSTE or DRX PET/CT scanners in 2 body segments. 2 radio-opaque markers (BBs) were attached to each patient on each leg. The BBs positions were placed in the overlap section of both segments. An automated software tool was developed to stitch the two segments and output a single WB DICOM image set. The software calculates the relative positions of the BBs on the 2 segments of the CT image set and a rigid transformation that matches the BBs between the two segments is generated to merge the two segments into one WB image. This process is facilitated by splitting the lower segment into two leg parts since patients might have moved between the different imaging segments. The same transformation was then applied to the PET image set. Decay correction was applied to the second segment to compensate for the difference in scan time between the segments. A single DICOM WB PET/CT image set was then generated for interpretation on two different workstations. The results showed that the stitched images were readable by the two workstations with good image quality at the stitching sight. The fusion of PET & CT on stitched images was performed as regular PET/CT scans. The software tool allows the merging of two segments of a WB PET/CT scan with minimal user interaction.