OBJECTIVE ASSESSMENT OF SUCCESSFUL SMILE REANIMATION: A Multi-Modal Approach

Abstract

Normal smiling is a complex phenomenon comprising both a social, communicative function which is voluntary, and a spontaneous, emotional expression of joy which is innate. For those afflicted by facial palsy, loss of the ability to smile has a profound effect that is not only physical, but also psychological and social. Whilst surgical techniques to restore facial movements exist, presently there is no outcome measure that objectively captures all the domains that facial reanimation impacts.

The central tenet of this thesis was to objectively quantify the success of smile reanimation surgery to restore normal smiling in facial palsy using a multi-modal approach. Chapter 1 introduced the concepts of voluntary and spontaneous smiling discussing the evolution of surgical techniques for facial reanimation in people living with facial palsy. Chapter 2 reviewed the ability of existing clinical outcome grading scales to quantify outcomes of smile reanimation procedures by conducting a literature review. Chapter 3 determined the relative importance of voluntary and spontaneous smiling using a ranked preference methodology known as discrete choice experiment. Chapter 4 measured facial movements on the healthy and diseased sides of the face and quantify the formation of the nasolabial fold using state of the art facial landmark recognition tools. Chapter 5 automatically detected and quantified the ability of spontaneous smiles to express joy using dual machine learning algorithms.

The thesis demonstrated that objective assessments of the success of facial reanimation procedures utilising a multi-modal approach are possible. We now can measure the physical impact through objective forms of smile measurements and analysis of smile spontaneity, the psychological effect on the individual through patient reported outcomes measures, and the social effect through automated forms of layperson assessments.