Measuring Asymmetry in Facial Morphea via 3D Stereophotogrammetry

Published: 12 May 2022| Version 1 | DOI: 10.17632/scm3n5fxtw.1
Laila Abbas, Adrienne Joseph, Jennifer Day, Naomi Cole, Rami Hallac, Christopher Derderian, Heidi Jacobe


Supplemental Figure 1A-B: Stereolithography (STL) models of a patient with right sided hemifacial atrophy. Layer one is the untouched STL model (A). Layer two is the STL model with the unaffected side (patient’s left) mirrored over the mid-sagittal plane (B). Supplemental Figure 2: Qualitative Rating Flow Chart. Supplemental Figure 1 depicts the flow of the qualitative rating process and the inter- and intra-rater reliability for both experts and non-experts. All raters first reviewed a set of training 2D clinical photographs depicting mild, moderate, or severe disease, and then rated both 2D clinical images and 3D stereophotogrammetry as mild/moderate/severe. Raters repeated their ratings 4 weeks later to determine intra-rater reliability. Both experts and non-experts determined one final rating for each patient after discussion within the group, which was then compared between experts and non-experts. One final rating was assigned after discussion amongst the whole group for purposes of subsequent analysis. Supplemental Figure 3A-C: 3D Stereophotogrammetry can be used to detect asymmetry in facial morphea, often challenging to appreciate clinically and on traditional photographs alone. This figure depicts a 15-year-old male with craniofacial morphea through multiple angles in traditional clinical photographs (A) and 3D stereophotogrammetry-generated three-dimensional images (B) and heat map (C), demonstrating asymmetry that was not readily apparent on clinical examination. Often, asymmetry can be difficult to assess clinically but can be easier to appreciate when using 3D stereophotogrammetry. Supplemental Figure 4A-D: 3D Stereophotogrammetry of En Coup De Sabre Morphea and Parry Romberg Syndrome. 3D-Stereophotogrammetry-generated three-dimensional images (A,C) and heat maps (B,D) are provided for a 22-year-old male patient with linear morphea en coup de sabre (A,B), and a 14-year-old female patient with Parry Romberg Syndrome (C,D). Supplemental Figure 5A-H: Comparison of Asymmetry Values to LoSCAT Scores. PGA-D scores correlated to lower face asymmetry (cheek and mouth), but LoSDI scores do not correlate to asymmetry values of any facial subunit. Supplemental Figure 3 A-H demonstrates the correlation between the LoSDI and PGA-D to asymmetry values of different facial subunits. LoSDI scores did not correlate to asymmetry of any facial subunit (A-D), in contrast to the PGA-D, which correlated to mean cheek and mouth asymmetry (F, H). This emphasizes the weakness of the LoSDI in facial morphea. The LoSAI was excluded from this analysis it does not include tissue loss



University of Texas Southwestern Medical School


Dermatology, 3D Imaging, Autoimmune Disorder, Facial Skin, Morphea, Facial Representation