Project 463581

Radiotherapy is an essential part of cancer treatment. Although 50% of curative-intent cancer therapy involves radiotherapy, the availability of radiotherapy varies greatly as resource constraints are an ongoing challenge. We have shown that artificial intelligence (AI) can increase clinical capacity and shift valuable human resources to more critical needs. Most medical AI technologies do not reach routine clinical practice despite the promise of AI to enhance the value of care. AI is often not built to integrate with complex clinical workflows and is not designed to consider how clinicians with use AI results. Although many AI algorithms have performance approaching or surpassing humans from a technical point of view, logistical translation challenges represent a chasm from the ideal, simulated research environment to the clinical setting. The overall goal of the proposed research is to address the gap between technical development and clinical use of AI in medicine by capitalizing on our existing clinically deployed AI radiotherapy technology. We will develop and prospectively validate a fully automated and clinically integrated AI-only end-to-end radiotherapy workflow and we will transform patient care in radiotherapy, while gaining a deeper, universally applicable understanding of clinicians' perceptions of AI in medicine. This groundbreaking study will provide a framework for closing the gap between technical development and clinical use of AI in medicine, elucidating this important conundrum for the greater medical community interested in the use of AI to enhance patient care. Furthermore, expediting the radiotherapy process will have immediate and tangible impacts for cancer patients and our healthcare system: enabling "same-day" and adaptive treatment paradigms that reduce the cost for additional patient visits, improving the patient experience, and liberating expensive human expertise to address more complex tasks and to increase patient capacity.