3393 - Optimizing Radiation Oncology: Integrating Clinical Decision Support and Automated Treatment Planning for Enhanced Quality, Consistency and Efficiency

K. Lai¹, J. Corry², S. H. Baxi³, M. Burke⁴, M. V. Dreosti⁵, M. Ng⁶, P. C. OBrien⁷, T. Fox⁸, F. Hughes⁹, and M. Price⁹; ¹Lumonus, Adelaide, SA, Australia, ²GenesisCare St. Vincents Hospital, Melbourne, Australia, ³Genesis CancerCare WA, Wembley, WA, Australia, ⁴GenesisCare, Sydney, NSW, Australia, ⁵Genesis Cancer Care SA, Adelaide, SA, Australia, ⁶Genesis Care, Melbourne, VIC, Australia, ⁷Calvary Mater Newcastle Hospital, Newcastle, Australia, ⁸Lumonus, Atlanta, GA, ⁹Lumonus, Sydney, Australia

Purpose/Objective(s): This study investigates the integration of a cloud-native Radiation Oncology Clinical Decision Support (RO-CDS) system, Horizon, and an Automated Treatment Planning Framework in Australian Radiation Oncology centers. The aim is to evaluate the impact on standardization, consistent quality, and decreased variability, focusing on enhancing quality and efficiency within a standardized framework. Materials/

Methods: Horizon involves more than 178 radiation oncologists at 43 centers in Australia, utilizing evidence-based Care Pathways (CPs) developed collaboratively with references like NCCN, EVIQ, and NICE. In the Automated Treatment Planning Framework, in 44 centers in Australia (urban and rural), utilizes Eclipse and Monaco systems. The study assesses compliance to CPs across 60% of the most common cancers prescribed with integrated automated planning, usability (System Usability Scale), and clinical workflow efficiency. It also looks at examining how the framework impacts on planning time, clinician approval rates, and dosimetry-related errors, with a focus on quality and efficiency improvement.
Results: Since 2018, Horizon has generated over 190,000 prescriptions, achieving 97% compliance and an SUS score of 84.2, indicating excellent usability. Implementation led to a 20-minute decrease in documentation time and prescription times averaged 3 minutes per patient, highlighting efficiency and consistent quality. The Automated Treatment Planning Framework, initiated in August 2021, processed >40,000 plans with a >50% clinician approval rate for >66.6% average automation utilisation across >19 common tumour streams covering VMAT, IMRT, 3D CRT, and SBRT/SRS techniques. It reduced the average time between target volume delineation and plan approval by 25%, from 1.6 to 1.2 days. Dosimetrist productivity increased by 40%, with an 85-minute time saving per plan. A 72% reduction in dosimetry-related errors contributed to enhanced quality and efficiency.
Conclusion: Integration of Horizon RO-CDS and the Automated Treatment Planning Framework successfully advances standardization, consistent quality, and efficiency in radiation oncology. Emphasizing adherence to evidence-based CPs, excellent usability, reduced planning time, and a significant decrease in dosimetry-related errors, this study aligns with physicians perspectives on quality improvement in radiation oncology practices. It offers valuable insights for ongoing excellence in patient care and treatment planning within and beyond Australian Radiation Oncology centers.