Click here to

Session: Therapy General ePoster Viewing [Return to Session]

Development of An Automated Pre-Treatment Plan Evaluation

J Rembish*, P Myers, C Kabat, N Bice, K Rasmussen, D Saenz, N Kirby, N Papanikolaou, S Stathakis, UT Health San Antonio MD Anderson Cancer Center, San Antonio, Texas


PO-GePV-T-156 (Sunday, 7/25/2021)   [Eastern Time (GMT-4)]

Purpose: To ensure high-quality care, radiation oncology teams conduct pre-treatment plan evaluations which review the machine treatment parameters, patient set-up instructions, prescriptions, and radiation dose distributions. The current process is performed manually, which consumes a great deal of time and introduces the risk of human errors. Automating pre-treatment plan evaluations can improve clinical efficiency and reduce errors.

Methods: Using Python, a web-based user-interface was created to automatically compare the data from the DICOM export and the Mosaiq SQL database. Various machine parameters, prescription details, and patient information are taken into consideration when cross-comparing between the two systems. The results are displayed side-by-side for each field being evaluated, and any discrepancies are highlighted for easy recognition. Additionally, the DVH is calculated and structures are assessed to determine if, and by how much, all constraints are being met. These scripts are publicly available through PyMedPhys.

Results: A data-transfer check has been developed to ensure a plan’s DICOM files and Mosaiq’s database information are consistent between general patient information and beam specific information. The comparison can be performed in a matter of seconds, making it significantly less time consuming than a manual comparison. Additionally, plan quality can be determined through DVH analysis and compared to an archive of patients.

Conclusion: The data transfer portion of a pre-treatment check can be automated and used to drastically reduce the amount of time required to complete the task while minimizing opportunity for human error. Implementing this into clinical workflow has the potential to increase efficiency while also increasing error detection sensitivity and improving plan quality.

Funding Support, Disclosures, and Conflict of Interest: CPRIT Pre-Doctoral Trainee (RP 170345)



    Quality Assurance, Computer Software, DICOM-RT


    Not Applicable / None Entered.

    Contact Email