Digital imaging analysis to assess scar phenotype

Brian J. Smith, Nichole Nidey, Steven F. Miller, Lina M. Moreno Uribe, Christian L. Baum, Grant S. Hamilton, George L. Wehby, Martine Dunnwald

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


In order to understand the link between the genetic background of patients and wound clinical outcomes, it is critical to have a reliable method to assess the phenotypic characteristics of healed wounds. In this study, we present a novel imaging method that provides reproducible, sensitive, and unbiased assessments of postsurgical scarring. We used this approach to investigate the possibility that genetic variants in orofacial clefting genes are associated with suboptimal healing. Red-green-blue digital images of postsurgical scars of 68 patients, following unilateral cleft lip repair, were captured using the 3dMD imaging system. Morphometric and colorimetric data of repaired regions of the philtrum and upper lip were acquired using ImageJ software, and the unaffected contralateral regions were used as patient-specific controls. Repeatability of the method was high with intraclass correlation coefficient score > 0.8. This method detected a very significant difference in all three colors, and for all patients, between the scarred and the contralateral unaffected philtrum (p ranging from 1.20-05 to 1.95-14). Physicians' clinical outcome ratings from the same images showed high interobserver variability (overall Pearson coefficient = 0.49) as well as low correlation with digital image analysis results. Finally, we identified genetic variants in TGFB3 and ARHGAP29 associated with suboptimal healing outcome.

Original languageEnglish (US)
Pages (from-to)228-238
Number of pages11
JournalWound Repair and Regeneration
Issue number2
StatePublished - 2014

ASJC Scopus subject areas

  • Surgery
  • Dermatology


Dive into the research topics of 'Digital imaging analysis to assess scar phenotype'. Together they form a unique fingerprint.

Cite this