MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Improving Plan Quality Through Automation of Treatment Planning Processes Using Scripting in RayStation

G Jarry*, M Ayles, M Brunet-Benkhoucha, D Martin, Hopital Maisonneuve-Rosemont, Montreal, CA,

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose:
In 2019, our center implemented automation of our planning processes using a graphical user interface (GUI) and a centralized database. The automation of treatment planning has lead to a global increase in the dosimetric quality of treatment plans.

Methods:
Planning processes are automated from the CT scan to the treatment room. A Python-based GUI was developed to guide the technologists, the radiation oncologists and the planners through the planning steps. These tools are triggered in the planning system (RayStation, RaySearch). The centralized database is used to store standard contour info, streamline transfer of information and collect patient-specific treatment dose statistics.
Treatment targets and dosimetric clinical goals are defined by the radiation oncologists and saved into the centralized database. This eliminates any possible confusion regarding target definition, fractionation, dose levels, and clinical goals. Scripts use this information to generate optimization contours, define treatment beams, create optimization objectives and optimize the plan. An iterative optimization process evaluates PTV coverage, organ at risk doses and global maximum dose to adjust the dose objectives and reach the optimal treatment plan.

Results:
Automation is now used for all of our IMRT/VMAT planning techniques and has been shown to reduce planning time and allow organ at risk dose reduction in most plans. In the case of prostate plans, full automation has allowed us to reduce rectal dose and hence more than double the number of plans receiving the full prescription dose. By reducing rectal dose, it has also allowed an increase in the number of patients eligible for hypofractionation by a factor of 1.5. For head and neck plans, the number of plans with parotids gland receiving less than 2600 cGy was increased by 20%.

Conclusion:
Automation has allowed us to reduce planning time, standardise our treatment plans and increase their dosimetric quality.

Keywords

Optimization, Treatment Planning

Taxonomy

TH- External Beam- Photons: Treatment planning using machine learning/Knowledge Based Planning/automation

Contact Email