Authors: I Frosio M Spadea E De Momi M Riboldi G Baroni G Ferrigno R Orecchia A Pedotti
Publish Date: 2006/02/16
Volume: 34, Issue: 4, Pages: 677-
Abstract
Patient setup optimization is required in breastcancer radiotherapy to fill the accuracy gap between personalized treatment planning and uncertainties in the irradiation setup Optoelectronic systems allow implementing automatic procedures to minimize the positional mismatches of lightreflecting markers located on the patient surface with respect to a corresponding reference configuration The same systems are used to detect the position of the irradiated body surface by means of laser spots patient setup is then corrected by matching the control points onto a CT based reference model through surface registration algorithms In this paper a nondeterministic approach based on Artificial Neural Networks is proposed for the automatic realtime verification of geometrical setup of breast irradiation Unlike iterative surface registration methods no passive fiducials are used and true realtime performance is obtained Moreover the nondeterministic modeling performed by the neural algorithm minimizes sensitivity to intrafractional and interfractional nonrigid motion of the breast The technique was validated through simulated activities by using reference CT data acquired on four subjects Results show that the procedure is able to detect and reduce simulated setup errors and revealed high reliability in patient position correction even when the surface deformation is included in testing conditionsThe deformation model was derived by analyzing multiple CT scans acquired in the free breathing FB and deep inspiration breath hold DIBH conditions of one patient treated for left breast carcinoma A geometrical surface deformation model was implemented in order to include the effects of patients breathing in the proposed ANN method for patient positioning correctionFB CT slice Z = 0 of the breast cancer patient left panel with the registered FB and DIBH surface contours superimposed white dotted lines In the right panel the mean ± 1SD values evaluated over the whole slice dataset of radius ρ black lines and of the radial difference Δρ grey lines are represented as a function of azimuth coordinate θ
Keywords: