Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Automated Facial Image Analysis: Detecting Improvement in Abnormal Facial Movement After Treatment With Botulinum Toxin A

Rogers, Carolyn R. BS*; Schmidt, Karen L. PhD; VanSwearingen, Jessie M. PhD, PT; Cohn, Jeffrey F. PhD§∥; Wachtman, Galen S. MD*; Manders, Ernest K. MD*; Deleyiannis, Frederic W.-B. MD, MPhil, MPH*

doi: 10.1097/
Advances in Plastic Surgery

The purpose of this study was to evaluate the ability of Automated Facial Image Analysis (AFA) to detect changes in facial motion after Botox injections in patients with facial nerve disorders accompanied by abnormal muscle activity. Eight subjects received Botox for oral to ocular synkinesis (n = 6), ocular to oral synkinesis (n = 1), and/or depressor anguli oris overactivity (n = 3). Subjects were video-recorded during 2 directed facial action tasks before and after Botox treatment. AFA measurement and Facial Grading System (FGS) scores were used to evaluate the effects of Botox. After Botox, AFA detected a decrease in abnormal movements of the eyelids in all patients with oral to ocular synkinesis, a decrease in oral commissure movement for the patients with ocular to oral synkinesis, and an increase in oral commissure movement in all patients with depressor overactivity. The FGS scores failed to demonstrate any change in facial movement for the case of ocular to oral synkinesis and for 2 cases of depressor overactivity. AFA enables recognition of subtle changes in facial movement that may not be adequately measured by observer based ratings of facial function.

Eight patients with longstanding facial paralysis and abnormal muscle activity were treated with Botox injections. Automated Facial Image Analysis (a software program) proved superior to the Facial Grading System for detecting changes in facial movements following treatment.

From the *Division of Plastic and Reconstructive Surgery, †Department of Psychiatry, ‡Department of Physical Therapy, and §Department of Psychology, University of Pittsburgh, Pittsburgh, PA; and the ∥Robotics Institute, Carnegie Mellon University, Pittsburgh, PA.

Received July 13, 2006 and accepted for publication October 10, 2006.

Financial Disclosure: None of the authors have received or will receive compensation, financial or otherwise, for their use of Botox (Allergan Corp) or Automated Facial Image Analysis (proprietary software) in the current study.

Automated Facial Image Analysis was developed using NIH grant #MH 51435 to Dr. Jeffrey Cohn in collaboration with Dr. Takeo Kanade of the department of computer science at Carnegie Mellon University.

This study was reviewed and approved by the institutional review board of the University of Pittsburgh.

Reprints: Frederic W.-B. Deleyiannis, MD, MPhil, MPH, Suite 6B, Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261. E-mail:

© 2007 Lippincott Williams & Wilkins, Inc.