An Investigation of Using Extracellular Volume Fraction Derived From Dual-Energy CT For pancreatic Tumor Delineation
G Godfrey1*, G Noid1, P Knechtges1, W Hall1, J Shah2, B Erickson1, X Li1, (1) Medical College of Wisconsin, Milwaukee, WI, (2) Siemens Healthineers, Durham, NC
Presentations
TU-A-202-1 (Tuesday, 7/12/2022) 7:30 AM - 8:30 AM [Eastern Time (GMT-4)]
Room 202
Purpose: Low extracellular volume fraction (ECVf) correlates with poor vasculature uptake and possibly identifies necrotic or hypoxic tissues in pancreatic tumors. This work investigates the feasibility of delineating pancreatic tumors using ECVf maps derived from contrast enhanced dual-energy CT (CE-DECT).
Methods: CE-DECT data acquired during radiotherapy simulation for 12 pancreatic cancer patients, using a dual-source DECT scanner, were analyzed. For each patient, an ECVf map of the pancreas was generated from the simultaneously acquired low and high energy DECT series during the late arterial contrast phase and was used to delineate a volume of interest (VECVf) in the following steps: (1) calibrating DECT iodine contrast with known iodine concentrations in a phantom; (2) constructing a patient-specific iodine map based on the iodine calibration; (3) computing the ECVf map using the patient-specific iodine map and the patient's hematocrit level; (4) creating a binary map between the 1st and 25th percentile of the ECVf map in the pancreas; and (5) delineating the VECVf based on the clusters identified by applying a connected components clustering algorithm. The obtained VECVf was compared to the gross tumor volume (GTV) using various metrics including positive predictive value (PPV), true positive rate (TPR), Dice similarity coefficient (DSC) and volumetric mean distance to agreement (MDA).
Results: In 7 out of 12 cases, the VECVf reasonably correlated to the GTV, with average values of PPV, TPR, DSC and MDA of 0.828, 0.490, 0.556, and 2.83 mm, respectively. The PPV were above 0.700 for the 7 cases. For the remaining 5 cases, the VECVf were substantially affected by other compounding factors, e.g., large cysts, dilate ducts, and thus were not correlated to the GTVs.
Conclusion: This proof of concept demonstrated the feasibility of using low ECVf regions derived from CE-DECT to locate pancreatic tumor volume for a subset of patients.
Funding Support, Disclosures, and Conflict of Interest: This research was funded in part by Siemens Healthineers, Durham, NC.
Keywords
Dual-energy Imaging, Perfusion Imaging, Treatment Planning
Taxonomy
IM- CT: Dual Energy and Spectral
Contact Email