Program Information
A Novel Web-Based Tool for Quantification of VMAT/IMRT Treatment Plan Quality
M Fan*, F DeBlois, K Sultanem, G Stroian, McGill University, Montreal, Quebec
MO-D-105-5 Monday 2:00PM - 2:55PM Room: 105Purpose: To develop a novel web-based tool for VMAT/IMRT treatment plan evaluation and to design a unique plan Quality Index (QI) quantifier to aid in decision-making for SRS, SBRT, and ENT treatment planning and evaluation
Methods: A high level Python web-framework, Django, is used to develop the web application. Django uses an SQL-like database for data storage and retrieval. The front-end of the web application is styled with CSS and written in HTML. Tumor site dependent evaluation templates for SRS, SBRT and ENT plans are created in collaboration with physicians at our institution. Previously approved treatment plans are imported into the web application to populate the database for analysis. With physician feedback, retrospective treatment plans are subjected to scoring algorithms to develop a plan QI.
Results: The web tool is currently deployed internally at our institution for VMAT treatment planners. The web site also serves as a portal for site-specific treatment planning instructions. Specific plan details are imported to an SQL-like database when treatment plans are pushed to the tool. The electronic nature of the database allows new retrospective studies to be easily conducted. A site specific QI is developed to quantify SRS, SBRT, and ENT plan quality. When the tool is used, QIs are generated automatically and a histogram of site specific QIs is produced. This information can then be used to evaluate the best candidate plan for treatment approval. The database can be used to query legacy plan specific details in order to extract the optimization objectives that best fit the anatomy and prescription doses from a new planning case.
Conclusion: The full implementation of the tool aims to standardize and unify the planning and evaluation of IMRT/VMAT techniques. Validation of the QI robustness will include correlation with clinical outcome and inter-institutional case studies.
Contact Email: