Robotic Surgical Systems and Methods for Utilizing 3D Point Cloud Analysis for Sensitive Region Avoidance Helping a Surgical Robot See and Steer Around Delicate Tissue

This patent describes a robot-assisted surgery system that combines tracking sensors with a vision camera and uses 3D point cloud analysis to identify and avoid sensitive regions during an operation. In simple terms, it gives the robot a more complete map of what is in front of it so it can keep its tools away from tissue that should not be touched.

The Problem

Surgical robots are precise, but they only know what their guidance system tells them. Traditional image-guided surgery relies on tracking the position of known objects such as a bone or a tool using a localiser. The trouble is that the real surgical field also contains things the system was never told about, such as soft tissue, retractors, a surgeon’s hand or other structures that drift into the workspace. If the robot’s model of the scene is incomplete, it can move toward something it should have avoided.

What This Invention Does

The system pairs a localiser, which tracks the precise pose of a known surgical object, with a vision device that captures live image data of the whole scene. The two data sources are merged into a single coordinate system. The system then assigns a first virtual object to the known surgical target and looks at the point cloud of measured points to find anything sitting outside that expected shape. Those leftover points, which represent unexpected or sensitive material, are grouped into a second virtual object the robot is told to steer around. The result is a robot that reasons not only about where its target is but also about everything else in the field that it should treat as off limits.

Key Features

• Sensor fusion. Localiser tracking data and camera image data are merged into one common coordinate system.
• Point cloud reasoning. The system analyses the three-dimensional cloud of measured points to separate expected anatomy from unexpected objects.
• Automatic sensitive-region flagging. Points that fall outside the known surgical object are grouped into a virtual no-go zone.
• Dynamic awareness. Because the vision device sees the live scene, the protected regions can reflect what is actually present, not just what was planned.
• Built for robotic control. The virtual objects feed directly into how the robotic manipulator is allowed to move.

Who Is Behind It

The applicant is Stryker Corporation, one of the world’s largest medical technology companies and a major name in surgical robotics, orthopaedics and operating-room equipment. The named inventor is Donald W. Malackowski. The application is a divisional of an earlier Stryker filing.

Why It Matters

Safety and trust are central to wider adoption of surgical robotics. A system that can detect and avoid tissue and objects it was not explicitly told about adds a layer of protection that mirrors the situational awareness of an experienced surgeon. For Stryker, protecting this approach in Australia supports its position in a fast-growing robotic surgery market across hospitals and surgical centres.

Related Concepts

• Robot-assisted surgery – the broad field this system belongs to.
• Point cloud – the 3D data structure the system analyses.
• Image-guided surgery – the navigation approach this invention extends.
• Computer vision – the technology that interprets the camera data.
• Sensor fusion – the merging of tracking and vision data into one model.

---

AU 2026201897 was published in the Australian Official Journal of Patents on 2 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.