Surgical instrument detection: Datasets & Models

Manually counting and tracking surgical instruments is tedious, error-prone, and time-consuming, and a miscount can carry serious patient safety consequences such as retained foreign objects. Automating this with computer vision provides a consistent, always-on check that flags discrepancies faster than a manual tally and frees clinical staff to focus on the patient. In sterile processing, automated detection helps verify that trays are complete and correctly assembled before they reach the OR. The result is fewer counting errors, better documentation, and a reliable audit trail.

Teams gain faster, more consistent instrument counts and tray verification while reducing the manual burden on nursing and sterile processing staff. The workflow integrates with existing OR and sterile processing cameras, so there is no need to rip out current infrastructure, and it scales across multiple rooms and facilities. Trained detection models like RF-DETR deliver accurate localization even in cluttered scenes, and the pipeline can be retrained as new instrument types are added.

1. Upload an image: Drop any photo into the widget above to try surgical instrument detection for free, right on this page.
2. Preview instant results: The model detects scalpel, forceps, surgical scissors, clamp, retractor within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
3. Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train RF-DETR on your own images for production accuracy.
4. Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.