Predicting blood flow near hepatic tumors with ultrasound to improve RF ablations

University Of Colorado Denver, Aurora CO

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): [Aim The proposed research is intended to increase the liver cancer tumor eradication rate of Radiofrequency (RF) ablation and increase the number of candidate patients. Relevance The local tumor eradication rate for patients with primary or secondary malignant liver tumors smaller than 3cm who are treated using Radiofrequency (RF) ablation is over 83%; unfortunately, the majority of patients have tumors larger than 3cm. The success rate of local tumor eradication of larger tumors falls to less than 50% due to unpredictable variation in the size of the thermal injury created by RF devices between different patients and even within the same patient. The success rates are limited by local blood flow carrying away the heat applied during treatment which can result in some tumor cells not reaching the cell death temperature of 55¿C and an undesired continuation and propagation of the disease. Objective The proposed research develops a new method of quantifying the amount of blood flow within a Region of Interest (ROI) surrounding a tumor using 3D ultrasound imagery. The flow information will be evaluated as a predictor of RF ablation device performance for use by clinicians during ablation planning to ensure that all of the tumor cells are destroyed and a minimal amount of healthy tissue is disturbed (critical for patients with cirrhosis). Methods Software based algorithms will be developed to reconstruct patient specific liver vessel geometry from 3D ultrasound Imagery. The associated Doppler color/velocity data from within the patient vessels will be combined with angle corrected vessel geometry to establish a Flow Influence Index within the ROI set around the tumor by the clinician. The index will provide an indication of the amount and direction of local blood flow. The algorithm function will be verified with an anatomically correct liver flow model of known geometry over a range of normal and pathological flow rates. The predictive capability of the algorithms will be validated by planning and performing test ablations on discarded cattle livers and comparing the planned and actual ablation sizes.]