Click here to

Session: Multi-Disciplinary: Research Tools and Databases [Return to Session]

A Clinically Implemented Watchdog Program to Streamline Adaptive LINAC-Based Head and Neck Radiotherapy

M Aristophanous*, E Aliotta, A Caringi, B Rochford, N Allgood, P Zhang, Y Hu, P Zhang, L Cervino, Memorial Sloan Kettering Cancer Center, New York, NY


TU-IePD-TRACK 4-1 (Tuesday, 7/27/2021) 12:30 PM - 1:00 PM [Eastern Time (GMT-4)]

Purpose: We have developed and implemented an adaptive radiotherapy program, enabled by a watchdog system that detects and alerts the clinical team of anatomical changes that may require plan adaptation. In this study we report our initial clinical experience.

Methods: Nineteen patients with cancer of the oropharynx and hypopharynx were studied. Patients that received definitive radiotherapy with no resection were included. The patients were selected at the time of planning and were scheduled on an in-house software program called AWARE (Automated Watchdog for Adaptive Radiotherapy Environment). The AWARE software automatically identifies weekly CBCTs for the scheduled patients, deforms tracked structures to the new anatomy, obtains volumetric information of these structures and alerts the adaptive team of significant changes via a daily morning email. The changes are then reviewed for accuracy and discussed with the treating physician who decides whether the plan needs to be adapted. For this initial study the primary and nodal GTVs, and the parotids were tracked.

Results: The AWARE program was successfully deployed and used to monitor 19 patients with weekly CBCT. The primary GTV decreased on average by approximately 10% the first 3 weeks, while remaining largely stable weeks 4-7. The nodal GTV showed the largest average reduction (15%) after one week of treatment and continued to see reduction until the final week. We observed an average parotid volume change of -24% [-46% - +8%]. Of the 19 patients that were tracked, 5 received an adaptive plan. In the majority of adapted cases, adaptive plans resulted in significant reductions (>10%) in dose to critical OARs.

Conclusion: We successfully developed and deployed watchdog software that automatically monitors anatomical changes on CBCT and alerts the clinical team. The developed software can enable and streamline personalized treatments by finding the optimum time point for adapting treatment.



    Cone-beam CT, Deformation, Image Guidance


    IM/TH- Image Analysis (Single Modality or Multi-Modality): Computer-aided decision support systems (detection, diagnosis, risk prediction, staging, treatment response assessment/monitoring, prognosis prediction)

    Contact Email