Abstract

We introduce a novel method for modeling the natural anatomical variation of an organ along inter-patient and intra-patient axes using a bilinear model. Bilinear models are attractive for this purpose since one type of variation can be explored while the other is held constant. We apply our model to total 204 prostate shapes contoured from CT imagery of 12 different patients, and show that the learned bilinear models can fit both training and testing shapes accurately. We also demonstrate the superior performance of the bilinear model over a linear model with the same number of parameters in adapting to prostate shapes from a new patient acquired immediately prior to consecutive fractions of radiation therapy.

Notes

Poster presentation at the 2006 Thrust R2D Image Understanding and Sensor Fusion Methods Conference

Keywords

Bilinear Model, radiation therapy, prostate

Subject Categories

Diagnostic imaging

Disciplines

Biomedical Engineering and Bioengineering

Publisher

Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS)

Publication Date

2006

Rights Holder

Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS)



Click button above to open, or right-click to save.

Share

COinS