الفهرس 
A comprehensive assessment of facial paralysis based on machine learning techniques
SummaryOnly 14 pages are availabe for public view
 |  | from | 148 |  |  |
| | | from | 148 | | |
Abstract
Quantitative grading of facial paralysis (FP) is essential to evaluate the severity and to track improvement of the condition following treatment. This work includes three research studies. First, evaluating the performance of certain facial muscles based on surface electromyography. Second, using machine learning algorithms to classify six facial functions based on the 3D facial landmarks and Facial Animation Units (FAUs) captured by the Kinect V2 sensor. Third, assessment and classification of the severity of FP based on symmetry analysis and evaluation of facial movements. The developed FP grading system is fast, easy to use, non-invasive, low cost, quantitative, and automated.