Click here to

Session: Data Science Autoplanning and Autosegmentation [Return to Session]

An Integrated Reporting Dataset for Investigating Programmatic Modifications of Radiotherapy Treatment Planning Data

M Schmidt1*, P Szentivanyi2, F Reynoso1, N Knutson1, G Hugo1, E Sajo3, J Labrash1, B Sun1, P Zygmanski4, M Jandel3, A Price1, (1) Washington University School of Medicine in St. Louis, Saint Louis, MO, (2) Varian Medical Systems, (3) University of Massachusetts Lowell, Lowell, MA,(4) Brigham & Women's Hospital, Boston, MA

Presentations

SU-H430-IePD-F5-1 (Sunday, 7/10/2022) 4:30 PM - 5:00 PM [Eastern Time (GMT-4)]

Exhibit Hall | Forum 5

Purpose: With powerful application programming interface (API) features integrated into the radiation therapy Treatment Planning System (TPS), treatment planning automation, or programmatic modification of treatment planning data, has become a part of clinical routine in many institutions. The purpose of this work is to develop a mechanism for reporting changes in patient treatment planning data performed by software automation routines.

Methods: A SQL Server Reporting Services (SSRS) reporting service integrated with the Oncology Information System (OIS) allows for the creation of custom datasets for report generation. A custom dataset, compiled SSRS query, was generated in Microsoft Report Builder within the Aria Unified Reporting Application (Varian Medical Systems, Palo Alto, CA) to extract structure sets and treatment plans which had been modified by an automation routine. This dataset is then utilized to generate a daily report to be reviewed for auditing patient modified with custom software applications. The report was aggregated over 2 years and 6 months, compared to the output of direct SQL query, and institutional automation use is reported.

Results: Both the SQL query and automation report compiled 2,362 modification events applied to 330 unique patients through the validation and use over 57 unique versions of 15 API assisted scripts. Automation events were recorded for automated treatment plan generation and modification (1,181), automated contouring and contour modification (510), and linear accelerator QA plan generation (669).

Conclusion: Automated plan and structure generation and modification can be a powerful tool for clinical efficiency and standardization. The ability to audit patients which are modified by automated tools is recommended for any clinic utilizing automation.

Funding Support, Disclosures, and Conflict of Interest: Washington University receives research support from Siemens, Varian Medical Systems, and ViewRay. MCS reports honoraria and consulting fees from Varian Medical Systems, Inc. and consulting fees from Lifeline Software, Inc. outside of the supported work. GDH reports personal fees from Varian Medical Systems outside the scope of the present work.

Keywords

Treatment Planning, Data Acquisition

Taxonomy

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

Contact Email

Share: