GE HealthCare and Mass General Brigham are building an AI tool to cut through the chaos of radiation therapy planning

Importing treatment data for a single cancer patient can require up to 17 manual steps. That figure alone tells you something important about where radiation oncology stands today, and why the field is ripe for a different approach. GE HealthCare and Mass General Brigham announced this week that they are working together to develop a generative AI tool designed to pull structured and unstructured clinical information, notes, images, and treatment plans, into one accessible interface that care teams can actually query in plain language.

The announcement was made in Chicago on September 23, 2026. The research is focused on radiation oncology workflows, where the problem is not a lack of data but an excess of it, scattered across systems that rarely communicate. Nearly 60% of cancer patients receive radiation therapy at some point in their care, yet the planning process remains deeply fragmented. One study of more than 500 radiation oncology cases found over 44 variations in treatment-planning workflows. That kind of variability is driven largely by unstructured information, clinical notes, free-text reports, that clinicians must read, interpret, and act on without much technical support.

How will it work?

The proposed tool would be integrated into GE HealthCare's existing Intelligent Radiation Therapy platform, known as iRT. That platform already connects workflow tools and reduces manual handoffs between care teams. The new AI layer would go further by allowing clinicians to ask direct questions about a patient's treatment journey and receive a synthesized, summarised view of the relevant information. The research teams will explore how the tool handles:

• Structured clinical data from electronic records and imaging systems
• Unstructured content such as clinical notes, physician reports, and treatment summaries
• Cross-modality information spanning MR, CT, and theranostics workflows
• Natural language queries from radiation oncology clinicians at the point of care

The goal is not to replace clinical judgment. It is to make the information needed for that judgment easier to find, faster to process, and more complete at the moment decisions are being made.

Why does it matter?

Time is a clinical variable in oncology. The previous collaboration between GE HealthCare and Massachusetts General Hospital, which formed the technical foundation of iRT, helped reduce the time from patient intake to the start of treatment from as long as 30 days down to eight days. That reduction was achieved by connecting fragmented workflows. This new research asks what happens when you add AI-driven information retrieval on top of that connected infrastructure.

For health systems across the GCC, where national cancer strategies are accelerating under Saudi Vision 2030 and the UAE's National Cancer Strategy, the implications are direct. Regional oncology centres are scaling rapidly, often recruiting internationally trained staff who must orient quickly to local patient populations and existing systems. A tool that surfaces patient-specific information on demand has practical value in any high-volume, high-complexity setting. And because iRT is already deployed across hospitals worldwide, any validated AI capabilities developed through this research could reach GCC institutions without requiring an entirely new technology stack.

The context

This collaboration sits within a broader shift in how health technology companies are thinking about AI. The focus has moved away from isolated diagnostic algorithms toward tools that work inside existing clinical workflows, ones that reduce friction rather than add another system to manage. GE HealthCare's approach here is consistent with that direction. Rather than building a standalone AI product, the company is expanding a platform that oncology teams already use.

John Wolfgang, a principal investigator on the research and clinical physicist at Mass General Brigham, described the ambition clearly: making complex treatment information easier to access, understand, and act on. Sam Kandala, General Manager for Therapy Guidance at GE HealthCare, framed it as a progression from solving workflow problems to helping clinicians make better use of the information already available to them. Both positions reflect what health systems, including those investing heavily in digital health infrastructure across the Gulf, are increasingly asking from their technology partners: not novelty, but tangible clinical usefulness.

The research findings, when they emerge, will be worth watching closely. If the tool performs as intended, it could change the daily reality of radiation oncology teams working under significant time and resource pressure, wherever they are.