Abstract
Statistical decision theory has been applied to the treatment planning decision of radiation therapy. The decision involves the choice of parameters which determine the radiation dose distribution. To choose among dose distributions requires a decision rule which reflects the uncertainty of possible outcomes for any specific dose distribution and the various risks associated with each outcome. A relative gravity or morbidity is assigned to each possible complication of treatment. In this study, recurrence of the disease in the irradiated volume is considered as a complication. Dose response functions relating the probability of complication to the dose distribution inside the organ at risk have been developed. It will be shown that a decision rule resulting from the application of decision theory to radiotherapy treatment is to choose the dose distribution for which <U> is a maximum where
[Formula: see text]
with wi = relative morbidity of complication i on a scale from zero to one, and Pi = the probability of occurrence of complication i for the dose distribution under consideration.
Full text
PDF
Selected References
These references are in PubMed. This may not be the complete list of references from this article.
- Cohen L. A cell population kinetic model for fractionated radiation therapy. I. Normal tissues. Radiology. 1971 Nov;101(2):419–427. doi: 10.1148/101.2.419. [DOI] [PubMed] [Google Scholar]
- McDonald S. C., Rubin P. Optimization of external beam radiation therapy. Int J Radiat Oncol Biol Phys. 1977 Mar-Apr;2(3-4):307–317. doi: 10.1016/0360-3016(77)90089-x. [DOI] [PubMed] [Google Scholar]
- Redpath A. T., Vickery B. L., Wright D. H. A new technique for radiotherapy planning using quadratic programming. Phys Med Biol. 1976 Sep;21(5):781–791. doi: 10.1088/0031-9155/21/5/007. [DOI] [PubMed] [Google Scholar]
- Van der Laarse R., Strackee J. Pseudo optimization of radiotherapy treatment planning. Br J Radiol. 1976 May;49(581):450–457. doi: 10.1259/0007-1285-49-581-450. [DOI] [PubMed] [Google Scholar]