Toward an Objective Outcome in Facial Reanimation Surgery: An Eyetracking Study
Thanapoom M. Boonipat1, Amjed Abu-Ghname, MD1, Ali Charaffadine, MD1, Kevin D. Fleming, PhD2, Mitchell Stotland, MD3, Samir Mardini, MD1.
1Mayo Clinic, ROCHESTER, MN, USA, 2Norwich University, Northfield, VT, USA, 3Sidra Medical, Doha, Qatar.
The availability of an objective outcome measure for facial reconstructive surgery remains elusive. Evaluations submitted by external raters or by patient self-report may be influenced by expert knowledge, emotional antecedent, or implicit attitude. These types of subjective ratings, or objective measures such as anthropometric analysis, may unreliably convey how one is perceived by others. We are interested in observers' instantaneous, reflexive responses to the human face, and how those instinctive responses relate to subjective judgment of a given face. We explored the visual markers that lead to differential perception of patients before and after facial reanimation surgery. By examining the early stages of visual processing that occur, we intended to measure changes in the focus of impression formation, thereby helping surgeons and their patients prioritize areas for reconstruction.
Pre and post operation (>3 months) photographs from 10 patients with unilateral facial nerve palsy who underwent cross facial nerve graft and free gracillis muscle transfer were obtained. Twenty lookzone regions were mapped onto each facial image, reflecting aesthetic units of the face. 40 observers examined each image while an infrared eye-tracking camera continuously recorded their eye movements. The observers were then asked to rate the image for character attributes (attractiveness, trustworthiness, sociability, healthy, and capability, 1-7 scale). Factorial ANOVA and student t-test analysis was performed to determine significance of differences between groups. RESULTS:
(i) The surgical intervention was found to decrease the observers' attention to the upper lip on the unaffected (non palsy) side, thus equalizing attention to both sides of the lips. (ii) The surgical intervention was found to significantly increase the character ratings for all five attributes compared to pre op controls: (sociable 3.2 to 3.6, trustworthy 3.42 to 3.61, attractive 2.96 to 3.2, health 3.37 to 3.54, p<0.05. (iii) For those pre-operative images of facial palsy, observers' attention was overwhelming drawn to the area of disproportion. (iv) Our eye tracking methodology clearly reflects a trend towards normalization of gaze attention following surgical intervention. This finding was associated with the improvement in character assessment of the images in the post-op cohort of images. CONCLUSION:
We provide data illustrating a novel and objective technique to evaluate the effect of reconstructive intervention for facial palsy. This information may be used to inform patients about how these areas of facial difference/aging are perceived, and the potential effect that surgical intervention may have on others' perception of them. This work may assist patients and their surgeons to more meaningfully focus their surgical decision-making priorities.
Back to 2019 Abstracts