Syeda Wafa e Zainab2025-09-102025-09-102025-06-07https://research.superior.edu.pk/handle/123456789/930Abstract Purpose: This study examines the efficacy of facial action unit (AU) analysis as an objective biomarker for detecting burnout and depression among dental graduates, addressing limitations of self-report measures in high-stress academic environments. Materials and Methods: A cross-sectional design was employed with 100 dental students (58% female; mean age=21.3±1.9 years). Participants completed the PHQ-9 for depression and MBI-Student Survey for burnout, while OpenFace 2.0 analyzed neutral facial images for AU4 (brow lowerer), AU1 (inner brow raiser), and facial asymmetry indices. Multiple regression analyses controlled for age and gender. Key Results: Strong correlation between AU4 intensity and burnout scores (r=0.42, p<0.01).Sadness Index predicted 38% of depression variance (R²=0.38, β=0.42, p<0.001).Combined facial metrics outperformed single predictors (ΔR²=0.12, p<0.01) Conclusions: Automated facial analysis demonstrates clinical potential as a supplementary screening tool, with AU4 and facial asymmetry serving as robust physiological markers of psychological distress in dental education settings. Keywords: affective computing; mental health screening; action units; academic stress; machine learning