MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel
  1. Home
  2. Meeting Program
  3. All
  4. Integrating Knowledge-Based Models for An Enhanced Iterative Automated Treatment Planning Process

Integrating Knowledge-Based Models for An Enhanced Iterative Automated Treatment Planning Process

M Vaccarelli1,2*, J Baker3, (1) Hofstra University, Hempstead, NY, (2) Northwell Health, Lake Success, NY, (3) Zucker School of Medicine at Hofstra/Northwell, Hempstead, NY

Presentations

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

Room: AAPM ePoster Library

Purpose:

Automated treatment planning through an iterative optimization approach that utilizes knowledge-based planning (KBP) can assist in yielding quality plans with time savings. We present a method to achieve enhanced results through combining both strategies.

Methods:

Varian’s Eclipse Scripting Application Programming Interface (ESAPI) and Rapid Plan version 15.6 are employed. A segmented low risk prostate CT image is input into the automated treatment planning workflow. Associated constraints are entered, and cumulative dose volume histogram (DVH) estimates are found. An optimization is first run based on initial organ at risk (OAR) constraints. A DVH is then calculated and assessed to determine if initial constraints are met. The DVH is further compared against the estimated DVH. If necessary, the objectives are adjusted and the process repeats itself.

Results:

The iterative automated treatment planning process we have designed consists of two main evaluation stages: optimization constraint modification and DVH comparison. First, if the DVH does not meet the planning goals, optimization parameters are adjusted. The priority, volume and/or the generalized Equivalent Uniform Dose (gEUD) alpha value are manipulated for each OAR constraint. After the DVH meets initial constraints, the estimated DVH is compared with the calculated DVH. The area under the curve is divided into segments to see if planning target volume (PTV) coverage or OAR sparing can be improved, while considering tradeoffs. The plan re-enters the calculating dose stage and subsequent DVH assessment until an acceptable dose distribution is attained.

Conclusion:

We present a workflow where an automated iterative planning approach is combined with KBP with the aim of achieving higher quality plans in less time. Future work will include analysis of auto-generated plans with clinically delivered treatment plans.

Keywords

Optimization, Treatment Planning, Dose Volume Histograms

Taxonomy

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

Contact Email

Additional Links