Artificial Intelligence in the Objective Comparison of Facial Rejuvenation Techniques
Methods/Technique: 32 patients who underwent facial rejuvenation surgery with concomitant procedures (Chart 1) between 01/01/2015 and 12/31/2017 were identified.10 patients underwent a SMAS plication facelift (Group A), 7 had a SMASectomy facelift (Group B), and 15 had a high SMAS facelift (Group C). Neutral repose (no expressed emotions) images pre- and post-operatively (average >3 months) were analyzed using the FaceReader. The software tracks 500 key locations on the face to accurately measure 28 action units and 7 cardinal emotions in each image. The action units and emotion are assessed for presence and intensity of functioning.
Results/Complications: Across the procedures only Group C experienced a change in emotional expression and action unit functioning (Graph 1). Post-operatively, 11/15 Group C patients’ experienced activation of the lip corner puller AU, increasing average intensity from 0% to 18.7%. This action unit pattern correlated with an average increase in detected happy emotion from 1.03% to 13.17% (p=0.008). Conversely, the average angry emotion detected decreased from 14.66% to 0.63% (p=0.032). Group A experienced a decrease in happiness by 0.84% and a decrease in anger by 6.87% (P>>0.1). Group B had an increase in happiness by 0.77% and an increase in anger by 1.91% (P>>0.1). Both Group A and B did not show any discernable action unit patterns.
Conclusion: This study provides the first proof of concept for the use of a machine learning software application to objectively compare aesthetic surgical outcomes in facial rejuvenation. This study demonstrates the objective increased efficacy in high SMAS facelift facial rejuvenation over other compared techniques. Future applications of this software include large patient population assessment of various facial surgeries to uncover which have the greatest impact on functioning and emotional expression.