Oncentra brachy optimization algorithms

Question 1

describe geometrical optimization

Answer 1

-wants dose uniformity within implant
o Recommended when four or more catheters, more than two planes, and/or an irregular volume implant is produced.
o Distance-based geometrical optimization is sensitive to dwell position separations along the catheter. Bigger separations correspond to further distance from the implant that a uniform dose distribution is produced.
o Does not need use-specified dose points (effectively uses source positions as dose points)

-can use distance based or volume based

Question 2

describe Points optimization

Answer 2

-optimizes dose at earlier defined points
o DTGR = dwell time gradient restriction. DTGR can be set to a value between 0 and 1000. High values of the DTGR correspond to smaller fluctuations between dwell times in neighbouring dwell positions.
o Recommended only for implants with only a few catheters

-can use distance based or volume based

Question 3

distance based for points or geometrical optimization

Answer 3

all catheters participate in the calculation. Relative weight of a particular dwell position is primarily determined by the neighboring active dwell positions in the same catheter (neighbouring dwells have the largest influence).
o Better for irregularly-spaced catheters

Question 4

volume based for points or geometrical optimization

Answer 4

o Better for regularly spaced catheters
only available if there is more than one catheter. The distance of active dwells in other catheters to the current dwell determines the relative weight of the current dwell. The catheter containing the current dwell is not considered. Therefore, the active dwell positions on the other catheters are the primary influence and this helps fill in spaces to cover the volume with the desired dose. Limits are imposed on total dwell times per catheter to make this work.

Question 5

describe graphical optimization

Answer 5

used to customize the shape of an isodose distribution by manually changing the isodose lines on the screen. You can also change individual dwell weights/times in the case explorer.
o Dragging the slider toward the “global” limit (away from “local”) increases the optimized area and the number of sources that will be affected.

Question 6

describe IPSA

Answer 6

inverse planning simulation annealing
-inverse optimization baed on OARs and dose objectives
-o Simulated annealing helps ensure that the iterative optimization procedure is not trapped in a local minimum by accepting some changes that result in a cost function increase
-the probability of accepting a cost increase decreases as the optimization progresses according to a pre-defined cooling schedule
- Stopping criterion can be the number iterations, final cost, of the difference between the 2-3 most promising solutions (if these are very similar, then stop)
o This method chooses which dwell positions to activate in addition to optimizing dwell times.
o Note that IPSA requires that the catheters be inside the target region by at least 5 mm (may need to create artificial optimization structure for this; i.e., create a PTV around your CTV).

Question 7

describe HIPO

Answer 7

hybrid inverse planning optimization
-combines both stochastic and deterministic algorithms. The stochastic algorithm is simulated annealing. The deterministic algorithm, called dose-volume histogram-based optimization (DVHO), optimizes three-dimensional dose distributions quickly by moving straight downhill (non-stochastic, gradient-based optimization) once it is in the advantageous region of the search space given by the stochastic algorithm

o For prostate treatments, HIPO uses the stochastic model to optimize catheter distribution (based on template holes projected on image, and contoured structures on the image), and a deterministic model (DVHO) to find optimal dwell times for the defined catheter configuration.

-user specifies sampling points and a margin for generatng surface dose points on OARs