Related Experiment Video
Updated: Oct 21, 2026
Quantitative Assessment Protocol for Facial Soft Tissue Volumetric Changes with Stereophotogrammetry
Published on: December 9, 2025
Quantifying Facial Rejuvenation after Deep Plane Facelift Using Convolutional Neural Network-Based Age Estimation:
Soner Karaali1, Aytaç Alten2, Burak Ergün Tatar3
1Department of Plastic, Reconstructive and Aesthetic Surgery, Private practice, Istanbul, Turkey.
Introduction:
To quantify perceived age reduction following deep plane face and neck lift surgery using five publicly available CNN models and evaluate the correlation between AI-estimated rejuvenation and FACE-Q Aesthetic Module outcomes.
Methods:
A retrospective cross-sectional study of 72 female patients undergoing deep plane face and neck lift surgery between 2020 and 2024 was conducted. Standardized photographs were analyzed using five CNN platforms: FacePlusPlus, Amazon Rekognition, Inferdo Face, howOld AI, and Facelytics. The arithmetic midpoint was used for the range-reporting models. A corrected and standardized age reduction metric was computed for each model and averaged into a composite score. The FACE-Q Aesthetic Module was administered at the 12-month follow-up. Spearman's rank correlation analysis was performed.
Results:
The composite standardized age reduction was 7.64 ± 3.44 years (15.67 ± 6.76%). Inter-model differences were not significant (Friedman test: χ2(4) = 9.23, p = 0.056). Patient-perceived age reduction correlated strongly with facial appearance (ρ = 0.61, p < 0.001), outcomes (ρ = 0.54), and psychological well-being (ρ = 0.44). AI-estimated rejuvenation showed weaker associations, with only the early life impact reaching significance (ρ = 0.25, p = 0.034).
Discussion:
CNN-based age estimation provides a reproducible and observer-independent metric for quantifying facial rejuvenation. AI-estimated and patient-perceived rejuvenation represent complementary dimensions of surgical outcomes, and their integration enables a more rigorous multidimensional evaluation framework.
Level Of Evidence Ii:
This journal requires authors to assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions for Authors www.springer.com/00266 .
