Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials
2017
Abstrakt
A highly efficient method is proposed to train and validate knowledge-based planning for use on radiation therapy clinical trials. Using plans from 86 patients who uderwent planning according to a multi-institutional cervical cancer trial we used protocol-specific dosimetric cutpoints to refine dose-volume histogram estimation models and implemented a program that automatically re-plans several candidate auto-planning routines across multiple patients.
After identification of the knowledge-based planning routine that best exemplified trial dosimetric aims, improved normal tissue sparing was observed across the sample.