Main Session
Sep
30
PQA 09 - Hematologic Malignancies, Health Services Research, Digital Health Innovation and Informatics
3662 - Harnessing AI for Sustainable Radiation Oncology: Transforming Brachytherapy Waste Management
Presenter(s)
Katie Lichter, MD MPH - University of California San Francisco, San Francisco, CA
K. Lichter1, G. Silva2, C. Phuong3, S. J. Liu2, L. Ni3, A. Smith2, J. Shin4, and I. C. J. Hsu5; 1University of California, San Francisco Department of Radiation Oncology, San Francisco, CA, 2UCSF, San Francisco, CA, 3Department of Radiation Oncology, University of California San Francisco, San Francisco, CA, 4Northeast Ohio Medical University, Rootstown, OH, 5University of California San Francisco, Department of Radiation Oncology, San Francisco, CA
Purpose/Objective(s):
The U.S. healthcare sector significantly contributes to greenhouse gas (GHG) emissions, with operating rooms accounting for approximately 30% of hospital waste. Within radiation oncology (RO), brachytherapy procedures therefore present an opportunity for physical waste reduction. Processing regulated medical waste (RMW) generates significantly higher greenhouse gas (GHG) emissions and incurs costs up to 10 times greater than standard landfill waste processing. Traditional waste audit protocols, while effective, are often time-consuming and resource-intensive. This study aims to explore the efficacy of artificial intelligence (AI) tools in optimizing brachytherapy waste management – reducing auditing time and waste volume – advancing RO’s evolving role in addressing climate and environmental challenges within health care.Materials/Methods:
A three-month waste audit was conducted within the brachytherapy suite of a tertiary academic RO department. Baseline data were collected manually across 20 brachytherapy cases using a standardized checklist to quantify RMW and landfill waste. An AI-driven platform employing machine learning (ML) algorithms (Zabble) was then introduced to analyze waste streams in real-time. This provided granular insights into waste stream patterns, facilitated cross-team communication, and supported the development of targeted waste reduction strategies. AI data informed the creation of a dynamic waste management system tailored to the RO context.Results:
RMW constituted 30-50% of total waste generated at baseline, averaging 6.16 kg per brachytherapy case (SD: 0.65 kg). Manual audits required an average of 4.5 minutes per case. The AI platform reduced audit time to 60 seconds per case while enhancing accuracy in waste tracking. The system identified improper waste segregation practices, leading to targeted staff education and behavioral interventions. As a result, inappropriate RMW disposal was eliminated in over 90% of cases, achieving cost savings of $4.22/kg and reducing landfill waste by over 20%. The AI-enabled platform also facilitated continuous waste monitoring, enabling the development of interventions to reduce single-use items and optimize supply utilization.Conclusion:
This study underscores the transformative potential of AI and ML technologies in promoting environmental sustainability within RO clinical practice. By streamlining waste management in brachytherapy operating rooms, an AI platform may not only reduce costs and environmental impact but may help to foster a culture of sustainability among healthcare professionals. RO continues to advance innovative care and contribute to global health initiatives; integrating advanced technologies into clinical practice may further enhance patient outcomes and drive secondary benefits, such as environmental sustainability and quality improvement.