PC54. How Does Brow Rotation Affect Emotional Expression as Measured Using Artificial Intelligence

Abstract

PURPOSE: We correlated manual brow rotations with emotional outcomes using artificial intelligence. METHODS: 52 preoperative photos of brow-lift patients’ brows were digitally rotated -20, -10, +10, and +20degrees in respect to the central axis (with midline rotation being positive) of their existing brow. The manipulated images and original pre/post-operative photos (totaling 312 images) were analyzed using FaceReader AI software. RESULTS: The primary emotion detected was neutral. The percentage of disgust in images detected increased with increasing positive degrees of brow rotation (2% disgust detected at-20degrees, 2.3% at-10degrees, 2.9% at neutral, 4% at+10degrees, and 5.2%at+20degrees), and sadness decreased with increasing positive degrees of brow rotation (27.6% sadness detected at -20degrees, 19.9% at -10degrees, 15.3% at neutral, 16.1% at +10degrees, and 15.9% at +20degrees). The intensity of the inner brow raiser decreased with increased positive brow rotation 0.088 at -20degrees, 0.011 at -10degrees, 0.038 at neutral, 0.03 at +10degrees, and 0.008 at +20degrees). Similarly, the intensity of the outer brow raiser increased with increased positive brow rotation (0.046 at -20degrees, 0.018 at -10degrees, 0.032 at neutral, 0.111 at +10degrees, and 0.139 at +20degrees). The facial action unit analysis also corresponded with the primary emotion analysis. The postoperative results show markedly more positive emotion expression (happy: 5% post operative compared <1% preoperative, including brow rotations, disgusted: 0.4% vs. 2-5%, sad: 1.1% vs. 15-27%) CONCLUSION: We demonstrated that increasing the degree of brow rotation correlated positively with the percentage of disgust and inversely with the percentage of sadness. Eye opening significantly changed emotions detected.