Using Facial Recognition Software to Quantify Perceived Age Reduction in Patients Undergoing Blepharoplasty.

Area of Science:

• Plastic Surgery
• Artificial Intelligence
• Medical Imaging Analysis

Background:

• Aging leads to periorbital changes like decreased skin elasticity and ptosis.
• Blepharoplasty is a common surgical solution for facial aging.
• Objective measures for blepharoplasty outcomes are limited.

Purpose of the Study:

• To evaluate convolutional neural networks (CNNs) for assessing perceived age changes post-blepharoplasty.
• To compare CNN-based age assessments with human evaluations.
• To identify factors influencing perceived age reduction after surgery.

Main Methods:

• Utilized pre- and postoperative blepharoplasty images from the American Society of Plastic Surgeons.
• Employed two CNN platforms (FacePlusPlus and Amazon Rekognition) to estimate patient ages.
• Used the Global Aesthetic Improvement Scale (GAIS) for human aesthetic assessments.
• Performed statistical analyses including paired t tests, linear regressions, and ANOVA.

Main Results:

• CNNs estimated an average perceived age reduction of 3.2 years post-blepharoplasty.
• Perceived age reduction correlated positively with GAIS scores (r = 0.33, P < 0.05).
• Greater age reduction was observed in patients who appeared older than their true age preoperatively.

Conclusions:

• CNNs can objectively quantify perceived age reduction following blepharoplasty.
• CNN results align with human aesthetic evaluations.
• CNNs show potential as objective tools for assessing surgical outcomes and managing patient expectations.