MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Computer Aided Verification of Radiotherapy Treatment Plans

JK Bowles1*, R Munbodh2 , (1) University of St Andrews, St Andrews, Fife, UK, (2) Warren Alpert Medical School of Brown University, Providence, RI, USA

Presentations

(Sunday, 7/14/2019) 5:00 PM - 6:00 PM

Room: 301

Purpose: Quality assurance (QA) in radiation oncology entails a multi-step verification of complex, high-technology treatments involving an inter-disciplinary team and multi-vendor software and hardware. Here, we extend our work on the pre-treatment physics chart review (TPCR) using a modular approach and formal methods from computer science to automatically identify logical inconsistencies in a patient’s treatment plan and how they propagate.

Methods: The TPCR process was decomposed into modules, subprocesses and module-associated variables for input to the subprocesses. The checks in the subprocesses and their order were formalized as constraints in an activity workflow diagram. The TPCR process was converted into a Satisfiability Modulo Theory (SMT) problem and solvers were used to 1) detect and correct logical inconsistencies in the TPCR specification through an iterative learning approach and 2) detect logical inconsistencies in a proposed patient treatment plan automatically.

Results: The model for the TPCR process comprises 5 modules, 17 sub-processes and approximately 300 variables (~70 distinct). Modules comprise “Patient Manager�, “Treatment Planning System�, “Independent MU Calculation� and “Record and Verify�. Subprocesses include “Dose Prescription�, “Documents�, “CT Integrity�, “Anatomical Contours�, “Beam Configuration�, “Dose Calculation�, “3D Dose Distribution Quality�, “IMRT QA� and “Treatment Approval�. We previously formulated our model to detect inconsistencies in the dose prescription, treatment modality and dose distribution. We have now extended the model to detect inconsistencies in additional areas including CT imaging, beam configuration, dose calculation and IMRT QA. Testing indicated that the solver successfully detected all inconsistencies in the specification of the TPCR process and in radiotherapy treatment plans automatically.

Conclusion: This work confirms that SMT solvers from computer science hold promise for automating the TPCR process and formalizing complex QA processes in radiation oncology and possibly other areas. The formalization and automation of these processes may lead to improved patient safety and increased clinical efficiency.

Keywords

Quality Assurance, Treatment Planning, Radiation Therapy

Taxonomy

Not Applicable / None Entered.

Contact Email