Visualization for Ultrasound Supported Biopsy

The goal of the project is to prepare 3D visualization software for the thyroid biopsy operation.

The first prototype was made with TrakStar hardware and Unity software. Although the TrakStar transmits the position and the rotation correctly, it is prototyped on the camera-based system because it is affected from the magnetic field and metals.

Using ArUco and the OpenCV library, 3D positions and rotations of ArUco markers were detected. Calibration was prepared using two markers. Tracking software was designed by attaching a marker to the ultrasonic transducer and the needle.

The TrakStar equipment could not be implemented in the clinical environment because it was affected from the magnetic field. Success was achieved with millimetric error in the test environment. The user-assisted calibration of the marker-based system resulted in a milimetric error.

Prepared with optimum calibrations and environment, the software prepared with ArUco markers has succeeded with the millimetric error rate. Because lines are drawn in two dimensions, the perspective problem is resolved by taking the position differences of the markers in the three-dimensional environment.