Click here to

Session: Multi-Disciplinary: Novel and Emerging Imaging Technologies in Radiation Therapy [Return to Session]

Developing Internal Liver Vascular Models for Radiation Therapy Assessment of Dose to Circulating Lymphocytes

C Correa Alfonso1*, J Withrow1, S Domal1, C Grassberger2,3, S Xing2,3, J Shin2,3, H Paganetti2,3, W Bolch1 (1) University of Florida, Gainesville, Florida, (2) Massachusetts General Hospital, Boston, MA, (3) Harvard Medical School, Boston, MA


WE-IePD-TRACK 3-3 (Wednesday, 7/28/2021) 5:30 PM - 6:00 PM [Eastern Time (GMT-4)]

Purpose: Lymphopenia is a side effect of radiation therapy that correlates with treatment outcome. The purpose of this work was the development of a model of the internal vasculature of the liver for the adult female and male (AF/AM) in order to calculate dose to the blood. Computational phantoms currently have no intra-organ vasculature that allows the calculation of dosimetric quantities in the blood.

Methods: Computer models of arterial, portal and hepatic vein trees were generated from optimization and physical principles using the Constrained-Constructive-Optimization (CCO) method. As liver has independent vasculature for each of the eight segments (following the Couinaud classification), AF/AM reference livers were divided into segments by matching the percentage of total liver volume to reported values of human liver. Next, the vessel generation algorithm was utilized to create independent vascular trees for all inflows (arterial and portal) and outflow (hepatic veins). Hemodynamic and geometric parameters of the main vessels were used as inputs. During the algorithm, pressure, blood flow, and radius of vessels in the tree are updated each time a new vessel is created and connected to the optimal bifurcation site.

Results: A closed vascular tree of approximately 6000 straight cylindrical non-intersecting pipes representing the hepatic artery, portal and hepatic veins was created independently for the AF/AM reference liver. While a straightforward comparison is difficult because of the variation in vascular structures, the morphology of our models shows similar trends to those observed in actual human liver vasculature.

Conclusion: We have developed computational models of human liver vasculature by implementing a vessel generation algorithm that creates binary trees representing realistic liver vasculature. Although the present model does not include the precapillary and capillary vasculature, or blood sinusoids, our model will be a useful tool to perform more realistic dose-calculations to circulating blood cells during radiation therapy.

Funding Support, Disclosures, and Conflict of Interest: This research is supported by NCI Grant R01 CA248901



    Blood Vessels, Blood Flow


    Not Applicable / None Entered.

    Contact Email