We propose a novel on-line functional imaging method with which tumor microenvironment can be quantified at the treatment unit. It is an X-ray contrast enhanced method based the (kv) cone beam CT (CBCT) platform and it promises to provide a surrogate measurement of tumor perfusion. The method was tested with 6 rabbits implanted with VX2 tumor. The technique involves a baseline CBCT followed by a contrast enhanced CBCT synchronized with contrast injection. The scan time for each CBCT is approximately 2 minutes during which a set of projections are acquired. The conventional reconstruction approach of contrast enhanced CBCT will give a single volumetric image with no temporal signature. On the other hand, the proposed technique makes use of a mathematical expression to parameterize the wash-in and wash-out behavior of contrast uptake in each voxel. Based on the projection data, the new method uses non-linear optimization to solve for the parameters describing the dynamic characteristics of injected contrast with which tumor perfusion can be derived. In our rabbit study, dynamic contrast enhanced images were reconstructed with the CBCT projections by the proposed method. Results show that the estimated contrast enhancement with the proposed CBCT method is in excellent agreement with the measured enhancement by CT. The proposed CBCT method is sensitive for detecting changes in the tumor microenvironment as a result of radiation treatment.
Keywords: Cancer; Computed tomography; Cone beam computed tomography; Image enhancement; Image reconstruction; Medical image contrast; Medical image reconstruction; Medical imaging; Optimization; Radiation treatment.
© 2008 American Association of Physicists in Medicine.