Datasets Pave the Way for High-precision Applications in Open Orthopedic Surgery

The lack of suitable datasets with 3D ground truth hinders an accurate surface mesh of the scene, which is essential for high-fidelity applications like marker-less surgical navigation and digitization of surgery.

Emma Most et al. evaluated 3D scanning, calibration of viewpoints for a set of high-resolution RGB images, and an optical method for scene registration. The results - generated during the course of PROFICIENCY - are highly promising to advance high-precision applications in surgical navigation and digitization.

Learn more here.

Figure: Comparison of our dataset acquisition method with SpineDepth. SpineDepth offers limited viewpoint diversity (2 camera poses), cadaver images that are unrealistic for surgery, a mean target registration error of 1.5 mm, and a median deviation between ground truth and measured anatomy of 2.4 mm. Our method allows for unlimited viewpoints (216 in

our experiments), and realistic images with a mean radial registration error of 0.35 mm.