According to the World Health Organization, cervical cancer affects nearly 600,000 women worldwide each year and causes around 340,000 deaths globally. Although brachytherapy remains one of the most effective treatments for cervical cancer, its use is often limited due to the time and expertise required to design personalised treatment plans.
A ‘One-Click’ AI Solution from UC San Diego
Now, researchers at the University of California San Diego have developed a groundbreaking “one-click” artificial intelligence (AI) tool that could revolutionise how clinicians plan brachytherapy treatments. The innovation promises to save physician time, reduce patient discomfort, and minimise the risk of human error. The research team published their findings in the journal Brachytherapy.
Why Current Planning Methods Fall Short
Brachytherapy is a form of internal radiation therapy that delivers radiation directly to a tumour while sparing surrounding healthy tissue. However, creating a customised plan for each patient typically takes an hour or more and depends heavily on the clinician’s experience. In many cases, patients must remain under sedation while doctors fine-tune treatment parameters.
Moreover, planning methods often vary between hospitals, leading to inconsistencies in treatment quality. As a result, access to optimal brachytherapy remains uneven, particularly in settings with limited resources.
Advanced Computing Meets Clinical Care
To address these challenges, researchers from UC San Diego’s School of Computing, Information and Data Sciences developed the new AI tool using advanced deep learning techniques and streamlined data processing. They trained and tested the system on hundreds of patient cases using the Voyager supercomputer at the San Diego Supercomputer Center (SDSC).
Encouragingly, the AI-generated plans matched the quality of those created by experienced clinicians, while requiring only a fraction of the time.
From Hours to Minutes
“We used U.S. National Science Foundation NAIRR Pilot allocations on SDSC’s Voyager system to develop this tool and integrated it into a widely used medical software platform, making it easy for clinicians to adopt,” said Lance Moore, an AI researcher with the UC San Diego School of Medicine’s Department of Radiation Medicine and Applied Sciences.
“With just one click, the system analyses a patient’s medical images and produces a high-quality, customised treatment plan in less than four minutes,” Moore added. “This approach can significantly reduce patient discomfort and lower the risk of human error.”
As reported by medicalxpress, Moore also acknowledged the contribution of Paul Rodriguez, a computational data scientist at SDSC, who collaborated on the model’s setup, testing, and benchmarking on the Voyager system.
Standardising Care and Expanding Access
Importantly, the researchers believe the tool could help standardise brachytherapy planning across hospitals, especially in clinics with fewer resources or limited specialist expertise. By automating routine planning tasks, clinicians could devote more time to refining treatment quality rather than racing against the clock.
Future Applications Beyond Cervical Cancer
Looking ahead, the team plans to adapt the AI tool for other cancers, including breast and prostate cancer. They also aim to make the technology easier to deploy across healthcare systems worldwide.
Ultimately, this AI-driven advancement has the potential to deliver faster, more consistent, and more equitable cancer treatment—bringing high-quality care to more patients than ever before.