Related Experiment Video
Updated: Jul 1, 2026
Quantitative Assessment Protocol for Facial Soft Tissue Volumetric Changes with Stereophotogrammetry
Published on: December 9, 2025
VisionMetric Suite for periocular measurement in ophthalmic plastic surgery: a preliminary single-center validation
Tengfei Wang1, Yunke Li2, Kun Wang1
1Department of Burns and Plastic Surgery, The Affiliated Hospital of Guizhou Medical University, Guiyang, China.
Background/Objectives:
Periocular plastic surgery lacks standardized, objective tools for preoperative planning and postoperative evaluation, currently relying on subjective assessment and manual measurements. This study aims to validate VisionMetric Suite, an artificial intelligence-based automated periocular measurement tool, and compare its performance with manual methods.
Methods:
We developed VisionMetric Suite based on the dlib library and shape estimation algorithms. The tool automatically measures 60 quantitative periocular parameters. The software was evaluated on 226 frontal facial photographs for recognition rate, processing speed, and user satisfaction. Thirty images were randomly selected to assess the agreement between automated and manual measurements and the repeatability of the AI tool using ICC and Bland-Altman analysis. Clinical utility was demonstrated through a case of left congenital ptosis.
Results:
The software achieved a recognition rate of 94.7% and an average processing time of 2.8 ± 0.4 s per image. User satisfaction among 10 physicians averaged 4.3/5.0. Agreement between automated and manual measurements showed a mean ICC of 0.94 (95% CI: 0.91-0.97), and repeatability of repeated AI measurements showed a mean ICC of 0.96 (95% CI: 0.92-0.98). Bland-Altman analysis confirmed these findings, and a representative clinical case illustrated the software's practical utility.
Conclusions:
This preliminary single-center validation study suggests that VisionMetric Suite can serve as a rapid, objective periocular measurement tool with good agreement compared to manual methods, representing an initial step toward automated assessment in ophthalmic plastic surgery.