Fig 1.
Schematic diagram of the workflow.
Schematic figure of the registration process during the different clinical and computational phases.
Fig 2.
Schematic figure of alignment by CLVs.
Fig 3.
Schematic figure of the Iterative Closest Point registration method.
In every iteration the optimizer was fine tuning the transformation parameters to find the best fit between the fixed and moving point sets. The moving point set was always the target of the transformation and the metric was designed to measure the fitness after the transformation. The iteration stopped if the optimizer could not find a better solution or if it reached a maximum number of iterations.
Fig 4.
Graphical representation of the axes.
The first set of axes for Case I. in 2D, on the level of the midsagittal plane, as the axes penetrate the plane. Fig 4A show the axes calculated as the first nonexcluded scan of the Bite0A position was used as the target of rotation for all nonexcluded scans of all opened bites. The opened scan of most centrally positioned axis (“x/+” marked symbols on the figures) was used to show the effect of error on the closed bites, thus all nonexcluded closed scans were used as the target of that chosen opened scan, shown on Fig 4B.
Fig 5.
The effect of the rms_error of the meshes on the daxis shows the EcD_ratio and its distribution.
Fig 6.
The effect of the rms_error of the meshes on the angular deviation of the axes.
Fig 7.
The amount of closure in degrees compared to the EcD_ratio.