(Wednesday, 8/1/2018) 9:30 AM - 10:00 AM
Room: Exhibit Hall | Forum 2
Purpose: To investigate the benefit of several types of dual-energy (DE) images (iodine maps, effective atomic number, and relative electron density) generated using the novel split-filter TwinBeam modality (Siemens Healthineers, Forchheim, Germany). Since pancreas tumor delineation using conventional CT is a challenge; this work evaluates the applicability of the DE images to aid in delineating pancreatic adenocarcinomas for radiation therapy applications.
Methods: Data was acquired for radiation treatment planning of pancreas cancer patients using the TwinBeam DECT system. Prior to imaging, all patients received iodine contrast to generate pancreatic and portal venous phase datasets. Iodine maps, effective atomic number, and relative electron density images were generated from the DECT data using DE postprocessing techniques of syngo.via (Siemens Healthineers, Forchheim, Germany) and compared to mixed 120 kVp-equivalent images and virtual monoenergetic images (VMIs). Analysis was performed using gross target volumes (GTVs) and region of interests (ROIs) within healthy pancreas tissue contoured by a radiation oncologist. The use of iterative reconstruction with DE images was also investigated.
Results: Iodine maps and effective atomic number images demonstrated a statistical difference between healthy pancreas tissue and pancreatic adenocarcinomas (p < 0.05), while relative electron density images did not. Iodine maps demonstrated 12% greater contrast to noise ratio (CNR) between pancreatic adenocarcinomas and healthy pancreas tissue compared to mixed 120 kVp-equivalent images. However, the CNR in the iodine maps was less than VMIs at 57 keV.
Conclusion: Iodine maps and effective atomic number images showed a significant difference between pancreatic adenocarcinoma and healthy pancreas tissue. Therefore, these DE images could aid in the delineation of pancreatic adenocarcinomas for radiation therapy applications by providing better qualitative and quantitative information of iodine enhancement compared to mixed 120 kVp-equivalent images.
Funding Support, Disclosures, and Conflict of Interest: Supported in part by a collaboration agreement with Siemens Medical Solutions
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
