Background Image
Table of Contents Table of Contents
Previous Page  496 / 1544 Next Page
Information
Show Menu
Previous Page 496 / 1544 Next Page
Page Background

ESTRO IMRT course 2015

IMRT optimization: algorithms&cost functions

Brussels October 5th, 2015

-19-

Static gantry, dMLC:

no distant local minima for `easy’ MLCs,

(e.g. sliding window)

proven algorithm for solution

Principles: The optimization problem for the

different IMRT techniques

Static gantry, step&shoot:

local minima possible, but good heuristics for

(‘step&shoot’ IMRT)

avoiding them (MLC sequencer)

(Notice: algorithm converges, heuristic produces a result)

Rotational-IMRT:

So many local solutions that it becomes impossible to

(VMAT, Rapid Arc ©…)

identify

the global one, bold heuristics for avoiding

the inevitable: getting trapped in a local minimum.