Oracle Health and Theator bring AI into the operating room

Surgeons spend years mastering their craft inside the operating room. Then they spend hours writing about it afterward, often from memory, often late. It is a broken workflow that costs hospitals money and, in some cases, introduces errors into the clinical record. Oracle Health and Theator think AI can fix it.

The two companies announced a collaboration on June 24, 2026, that will bring Theator's surgical intelligence platform to Oracle Health customers across the United States. The system records and analyzes surgical video in real time, cross-references it with electronic health record (EHR) data, and automatically generates a structured operative report by the time the surgeon walks out of the room. No dictation. No transcription. Just a report ready to review and sign.

The platform runs on Oracle Cloud Infrastructure (OCI), which handles the heavy computational load of processing high-definition video and running the AI models that interpret what is happening on screen moment by moment.

How will it work?

Theator's system uses computer vision and AI to watch a surgical procedure as it unfolds. It does not simply transcribe audio or log timestamps. It actually interprets the video: recognizing which step the surgeon is on, whether key safety checkpoints were reached, and which clinical events are worth documenting.

When the procedure ends, the system produces a report that flows directly into the Oracle Health EHR through Theator's Surgery-to-Text product. From there:

• Care teams can access clinical notes immediately after the procedure
• Quality reviewers get structured data they can benchmark against other cases
• Billing staff receive documentation that reflects the actual complexity of the surgery, reducing coding gaps

For hospitals already using Oracle Health, everything sits within the same security and compliance architecture they have today. There is no separate system to manage.

Why does it matter?

The accuracy problem with surgical documentation is well documented. A peer-reviewed study published in the Journal of the American College of Surgeons found that operative reports written from a surgeon's memory achieve only 72.8 percent accuracy. Theator's automated reports showed significantly higher accuracy in the same publication, with a p-value of 0.001, which is a strong statistical result.

The consequences of inaccurate documentation go beyond administrative headaches:

• Incomplete records can affect how future care teams interpret a patient's history
• Billing codes that do not reflect the true complexity of a procedure cost hospitals revenue
• Missing safety data makes it harder for health systems to spot patterns or improve outcomes

Theator has already analyzed more than 600,000 procedures across more than 150 procedure types, and the platform is active at major academic medical centers in the US and internationally, including Mayo Clinic and UHealth Miami.

The context

AI has moved quickly through most areas of medicine over the past few years. Ambient clinical documentation tools from companies like Nuance and Suki have already changed how doctors in clinics and hospital wards handle notes. The operating room has largely been left out of that shift, partly because of the technical complexity of processing surgical video and partly because the OR is a harder environment to standardize.

This partnership is a direct attempt to close that gap. Oracle Health is one of the largest EHR vendors in the country, which means a successful rollout here could put AI-generated surgical documentation in front of a significant number of hospitals quickly. The broader ambition, according to both companies, is to make surgical data as structured and searchable as any other clinical encounter, which would open the door to system-wide quality benchmarking and real-time safety monitoring across institutions. That is a meaningful shift in how surgical care could be evaluated and improved at scale.