Dermatologic Scar Assessment With Stereoscopic Imaging and Digital Three-Dimensional Models: A Validation Study.

Abstract

We evaluated a new handheld stereoscopic imaging system capable of visualizing scars with digital three-dimensional (3D) models and providing automated morphometric estimates. The objective was to validate the repeatability and accuracy of intra- and inter-investigator scan results. Engineered metal plates with depressed and elevated model scars (n = 72) were scanned six times by one investigator. In vivo hypertrophic and atrophic scars (n = 15) were scanned once by three investigators. The repeatability of morphometric estimates was assessed using coefficients of variation (CVs) to compare the variation among multiple scan results for both models and in vivo scars, with 0% reflecting a perfect match. Scar estimates from digital 3D reconstructions were compared with the known dimensions of physical model scars and with ruler measurements of in vivo scars. A total of 48 model scars and 12 in vivo scars were eligible for automated analyses with the imaging system's proprietary software. Intra-investigator scan results for the model scars were repeatable, with low variance for all parameters: volume, area, length, and depth/height (CV: 1.8-3.1%). By comparison, inter-investigator scans of real in vivo scars resulted in slightly higher median CVs (4.4-7.3%; P < 0.05). 3D model scar estimates correlated well with the known physical dimensions of model scars for all parameters (P < 0.001) and accurately reflected the measurements of in vivo scars (P < 0.001). The six in vivo scars situated on the chest and abdomen showed the highest inter-investigator variation, due to respiratory movement artifacts. Stereoscopic imaging of scars generates accurate and repeatable measurement estimates that show little intra- and inter-investigator-based assessment variation. The best results are achieved by minimizing subject movement. Lasers Surg. Med. © 2020 Wiley Periodicals LLC.