Quantitative facial analysis using emotrics in patients undergoing vestibular schwannoma surgery

Facial nerve dysfunction is a common complication after vestibular schwannoma (VS) resection. Traditional grading scales such as House-Brackmann (HB), Sunnybrook, and Fisch are subjective and prone to inter-rater variability. Emotrics, a computer vision-based tool, offers objective facial analysis, but its role in neurosurgical populations remains underexplored.

Patients undergoing VS surgery were prospectively enrolled in this study. Standardized frontal facial photographs were taken preoperatively, at discharge, and at three-month follow-up. Emotrics quantified facial symmetry and movement across static, dynamic, and synkinesis domains. Composite scores were generated and compared with clinical grading scales and patient characteristics.

Thirty-six patients (mean age 51.2 ± 12.3 years) were included. Emotrics detected significant improvement in dynamic function (mean score: 0.41 at discharge vs. 0.56 at 3 months, p < 0.01) and synkinesis (0.62 to 0.51, p < 0.05); static symmetry remained stable. Strongest correlations with clinical grading were found at discharge and 3 months. Discharge Emotrics total scores correlated significantly with HB (r = -0.883, p < 0.001), Sunnybrook (r = 0.892, p < 0.001), and Fisch (r = 0.883, p < 0.001). At 3 months, dynamic scores remained strongly associated with HB (r = -0.799), Sunnybrook and Fisch (r = 0.790; p = 0.001). Worse Emotrics outcomes were linked to older age (62.1 vs. 48.6 years, p = 0.04) and larger tumor volume (8.2 vs. 4.5 cm³, p = 0.03).

Emotrics offers objective assessment of facial nerve function, correlates well with clinical scales, and enhances evaluation of dynamic facial recovery. Its integration may refine postoperative monitoring and guide rehabilitation.