Software Based Video Analysis Of Functional Outcomes Of Face Transplantation
Sebastian Fischer, MD, Nicco Kretzdorn, MD, Anne Huang, BS, Muayyad Alhefzi, MD, Mario A. Aycart, MD, Harriet Kiwanuka, BS, Thet Su Win, MD, Yannick Diehm, BS, Ericka M. Bueno, PhD, Bohdan Pomahac, MD.
Brigham and Womens Hospital, Boston, MA, USA.
PURPOSE: Outcome assessment of face transplantation is mandatory to provide evidence about this live-giving surgery. Current methods, however, are imprecise or prone to subjectivity. Software-based video analysis allows for fast, objective and retrospective assessment of facial movements and expression of emotions.
METHODS: We recorded videos of each of our 7 face transplant recipients before as well as every 3 to 6 months after transplantation. In every video each patient performs the same sequence of facial movements. These videos were retrospectively analyzed by a customized software, which is capable for automatic tracking and detailed measurements of facial movements and expressions. Subsequently, measurements were compared to the same patient at different time points, to unaffected structures in case of partial face transplants as well as to a normal population.
RESULTS: After a mean follow up of 3.5 years every patient showed a significant improvement in facial movements compared to pre transplant time points. Significant improvements were reached after an average of 1.5 years after transplantation. In 3 patients with partial face transplants, extent and quality of facial movements of the transplanted part did not show any significant differences compared to the unaffected part 1 year after transplantation. 2 partial and 1 full face transplant achieved outcomes comparable to a normal population.
CONCLUSION: Software based video analysis provides the first assessment tool capable for objective, precise and reproducible analysis of facial movements and emotions. Of note, this software works with conventional camera and computer equipment making it worldwide applicable
Back to 2016 Joint Meeting Abstracts