Premium
Sci‐AM2 Sat ‐ 07: Development of inverse planning and limited angle CT reconstruction for cobalt‐60 tomotherapy
Author(s) -
Chng N,
Kerr A,
Rogers M,
Schreiner J
Publication year - 2005
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.2031057
Subject(s) - tomotherapy , iterative reconstruction , projection (relational algebra) , computer science , mathematical optimization , radiation treatment planning , medical physics , algebraic reconstruction technique , dosimetry , algorithm , mathematics , computer vision , nuclear medicine , medicine , radiology , radiation therapy
A significant amount of evidence exists to suggest that Cobalt‐60 can be used to deliver conformal treatments with intensity modulated tomotherapy and on‐line image guidance with megavoltage CT. The purpose of this paper is to describe our recent advances in developing the potential for Cobalt‐60 as a modality for tomotherapy. In particular, we have been advancing forward and inverse planning designed especially for our benchtop delivery configuration, on‐line image reconstruction for Cobalt‐60 CT, and approaches to dose reconstruction. Treatment planning is performed by dose optimization under implicit dose‐volume constraints for a slice using a gradient based inverse algorithm. Although a least‐squares framework is maintained in the explicit objective function for rapid convergence, regional weightings evolve over time to attempt to accommodate dose‐volume constraints. The selection of the appropriate relative region weightings in the objective function is handled internally to improve ease‐of‐use. Simulated plans under a variety of dose‐volume constraints reach reasonable solutions in a matter of minutes. The second area of focus has been in the creation of a CT method that is amenable to the deficiencies of image reconstruction under the inherently limited set of treatment data, while maintaining a high speed of reconstruction necessary for on‐line registration. A one‐step reconstruction method belonging to the class of algebraic reconstruction techniques (ART) is being tested and has shown promise in providing a compromise between the limited‐view qualities of iterative ART and the high reconstruction speed of filtered back‐projection.