10.1371/journal.pone.0234330.g002 Torsten Straßer Torsten Straßer Sandra Wagner Sandra Wagner Eberhart Zrenner Eberhart Zrenner Fig 2 - Public Library of Science 2020 ciliary muscle landmarks open-source software CilOCT OCT images Presbyopia ciliary muscle apex image processing software segmentation ciliary muscle biometry DICOM image distortion correction XML files ciliary muscle analysis 2020-06-09 17:27:22 Figure https://plos.figshare.com/articles/figure/Fig_2_-/12454661 <p>An illustrative example of the segmentation of the ciliary muscle: (a) The imported raw DICOM OCT image. (b) Manually placed guiding landmarks for the upper (green circles), lower (blue circles), and the nasal or temporal border (yellow circles) of the ciliary muscle, and a landmark in the iridocorneal angle (purple circle) as starting point of the 8-neighbor flood-fill algorithm (guiding landmarks size is enlarged for better visibility). (c) Maximum brightness gradients determined from the first rough polynomial spline fit of the landmarks. (d) Polynomial splines of the final segmentation. (e) Distortion correction based on Snell’s Law using the refractive indices of the different tissue layers. The incoming light rays of the infrared light beam and how they are refracted at the boundary between air and scleral-conjunctival tissue are exemplified as yellow lines perpendicular to the top of the images. Red and orange lines depict the refracted light rays. (f) Distortion corrected segmentation and determined biometric parameters scleral spur, ciliary muscle apex, perpendicular axis, and ciliary muscle area (all values in pixels).</p>