fig4
From: Deep learning-driven catheter tracking from bi-plane X-ray fluoroscopy of 3D printed heart phantoms
Figure 4. Validating 3D co-registration algorithm. (A) Image of 3D printed jig holding array of 50 metal spheres at various heights. (B) Image of fluoroscopy images at two angles and auto-detection of those spheres. (C) Graph of error for each sphere based on true value measured from 3D CAD file for bi-plane.
