GE HealthCare and Mass General Brigham are building a generative AI tool to help radiotherapy teams navigate fragmented patient data and speed up treatment.
The company has partnered with a Mass General Brigham research team to use generative AI to support radiation therapy teams, with the work focused on enabling radiation oncologists to query structured clinical data and unstructured information.
Generative AI could summarise a patient’s treatment journey to help clinicians navigate complex data, make decisions and deliver personalised care, GE HealthCare said.
John Wolfgang, a principal investigator on the research and a clinical physicist in the department of radiation oncology at Mass General Brigham, said the original work “demonstrated how technology can remove barriers in care delivery and help clinicians focus more of their time on patients while accelerating time to treatment.”
He said the team was “excited to advance this work to explore how AI could further support radiation oncology care teams by making complex treatment information easier to access, understand and act on.”
Nearly 60 per cent of people with cancer receive radiation therapy, yet planning and preparing that treatment has historically been fragmented across disconnected systems.
The length and complexity of treatment, plus the number of care teams involved, mean radiation oncology staff work across multiple systems and sources of information that “rarely connect”, GE HealthCare said.
The scale of the administrative burden is significant. Importing treatment data can take up to 17 manual steps, and one study of more than 500 radiation oncology cases identified over 44 variations in the treatment-planning workflow, according to the company.
That variability stems partly from unstructured information such as clinical notes, which are harder for care teams to search quickly, potentially delaying patient care.
Building on an existing tool
The new work builds on an earlier collaboration in which software developed at Massachusetts General Hospital became a component of GE HealthCare’s Intelligent Radiation Therapy (iRT) workflow management platform.
Based on analysis of more than 11,000 treatment plans at MGH, that integration is reported to have cut the time from patient intake to the start of treatment from as long as 30 days to eight days, though the company cautions results may vary between hospitals depending on workflows and patient populations.
The generative AI tool now being explored would combine structured clinical data with unstructured material such as clinical notes, images and treatment plans, letting radiotherapy teams ask questions and surface relevant details quickly. It could eventually be integrated into iRT and inform future capabilities across MR, CT and theranostics.
Sam Kandala, general manager for therapy guidance at GE HealthCare, said the collaboration was an opportunity to explore “how AI could help clinicians not only work more efficiently but also make better use of the information available to them as they personalise treatment for each patient.”
The work remains at the research stage, with no regulatory clearance or commercial product yet confirmed.
It comes as regulators scrutinise how generative AI should be governed in medical devices, with the US Food and Drug Administration recently seeking feedback on the technology, warning that it may introduce risks not seen with earlier forms of AI.
