Computerized Quantitative Analysis of Facial motions

Abstract

The scoring system is widely accepted as a simple and useful method for evaluating the degree of facial palsy, but there have been some problems in objectivity and reproducibility. This study investigated a new objective method of analyzing facial movements and examined the physiological range of asymmetry and differences between genders. Facial expression data input to an Apple Computer using Quick Time was analyzed by applying digital image editing techniques. Forty-six healthy volunteer subjects were analyzed utilizing this method. To quantify facial movements, several points on the face were identified to accurately monitor their movements. Twenty-four white marks were placed on the face, and 10 pictures in a series from resting to maximum movements were imported. The movements of each marker were expressed on a coordinate axis. Four voluntary facial movements (forehead wrinkling, eye closing, grinning and whistling) were examined. The trajectory of the marks were expressed numerically and changes in an area within a polygon produced by connecting all marked points with a line were measured. The ratio of the area of maximum movement to the area at rest was obtained and used for comparative investigation. The differences in the ratio between the left and right sides were less than 6% in any area and any movement. There were no significant differences due to gender. Some areas obtained by subdividing showed larger changes with facial movements than other areas. © 1994, The Society of Practical Otolaryngology. All rights reserved.