Comparing Deep Plane and Superficial Musculoaponeurotic System Facelifts Using Artificial Intelligence-Based Age Estimation.

Background: Although artificial intelligence (AI) is implemented in diagnostic medicine, few modalities exist to evaluate aesthetic outcomes. The literature demonstrates AI's ability to predict patient age mainly from frontal views, despite the importance of profile views in age perception. It is also uncertain how the predominant facelift techniques (deep plane versus SMAS) and adjunct procedures affect AI estimation. We aimed to investigate AI's ability to estimate age pre- and postoperatively using frontal and profile views, comparing outcomes across techniques and adjuncts. Methods: For each facelift technique in this retrospective study, Face++ AI age estimation software evaluated pre- and postoperative photographs of 100 female patients, identified from plastic surgery websites. Patients (N = 200) were grouped by facelift technique and adjunct procedures, and Microsoft Excel was implemented for data analysis. Results: Two hundred patients (100 deep plane, 100 SMAS; mean age 59.97 y) were analyzed using Face++ AI. Frontal images showed modest apparent age reduction (~2.5-2.8 y) with moderate accuracy, whereas profile images demonstrated substantial underestimation and high variability with a mean absolute error of 16-23 years. No significant differences in AI-estimated age reduction were detected between surgical approaches, and adjunct procedures did not significantly affect age estimates. Conclusions: This study uniquely evaluates Face ++ AI age estimation for SMAS and deep plane facelifts using both frontal and profile views within a single dataset. AI detected modest age reduction in frontal views but was inaccurate in profiles, and adjuncts had no significant effect, highlighting current AI limitations.