Incorporating Model Parameter Uncertainty into Prostate IMRT Treatment Planning
Abstract
IMRT has become one of the main tools for prostate cancer treatment. Current IMRT inverse planning is mainly performed using dose-based objective functions, which oversimplify the problem and ignore the useful biological properties of target and normal tissue. Although different biological model-based objective functions have been investigated in IMRT optimization, the parameters involved in the biological models are very coarse. The objective of this investigation is to establish a framework to consider model parameter uncertainties in prostate IMRT optimization. In order to implement this, a mathematical frameset was first established based on the estimation theory and statistical analysis. Biological model parameter uncertainties and clinical end point data were then incorporated into inverse treatment planning process and a clinically practicable inverse planning framework was established. 30 prostate cancer cases were studied using this technique. The results demonstrated that the proposed technique is capable of greatly improving the sensitive structure sparing without losses of target dose coverage and homogeneity. In addition, by including the model parameters uncertainties, we also implemented an algorithm to optimize the time-dose-fractionation for prostate cancer treatment. This investigation sheds important insight into the complex plan optimization and dose-time-fractionation problems and is valuable for improving prostate cancer patient care.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 2005
- Accession Number
- ADA439169
Entities
People
- David Y. Yang
Organizations
- Stanford University