Can AI Measure a Surgeon's Skill?

Researchers are linking surgical movements to recovery data, opening the door to objective performance insights in the operating room

13 Feb 2026

Artificial intelligence researchers are developing tools to measure surgical skill by analysing operating room video in granular detail, in an effort to move beyond traditional peer review and subjective assessment.

At the centre of the work is a platform known as Frame-to-Outcome. Rather than classifying procedures into broad stages such as “incision” or “closure”, the system reviews surgical footage frame by frame. It tracks fine instrument movements and examines how those patterns correlate with patient recovery and complications.

The aim is to identify techniques that consistently align with smoother healing or fewer adverse events. If such links can be established, researchers argue, they could help define best practice using measurable data rather than reputation or senior opinion alone.

For now, the technology remains largely confined to research settings. It is not a standard clinical tool, and most studies have been conducted under controlled conditions. Broader validation across hospitals, medical specialties and patient groups would be required before routine adoption.

In early studies focused on robot-assisted prostate surgery, the system performed almost as well as trained human reviewers in identifying specific surgical actions. Researchers also reported statistical associations between certain motion patterns and post-operative recovery markers. The findings are preliminary and do not establish causation, but they suggest a more detailed way to analyse performance.

The project reflects a wider push to apply artificial intelligence in healthcare, from automating clinical documentation to forecasting patient risk. Surgical workflow analysis has gained attention as robotic and minimally invasive systems generate large volumes of video and procedural data.

Significant questions remain. Regulatory frameworks for such tools are still evolving. Ethical concerns over data privacy, video ownership and informed consent have yet to be resolved. Commercial applications, including use in reimbursement or formal quality reporting, have not been defined.

Until these issues are addressed and evidence broadened, AI-driven surgical analytics is likely to remain a research exercise rather than a feature of routine care.