Evaluation and Severity Classification of Facial Paralysis using Salient Point Selection Algorithm

Region Based Parallel Hierarchy Convolutional Neural Network for Automatic Facial Nerve Paralysis Evaluation

“…Different with facial expression recognition, FNP assessment is to evaluate the intensity of facial deformation through the landmark distance features and spatio-temporal variation information involved in patient's facial salient areas. Some researchers also focused on the local symmetric features of facial palsy patients [33][34][35]. Liu et al [33] compared two sides of the face and represented the severity of the paralysis by calculating four ratios.…”

“…Liu et al [33] compared two sides of the face and represented the severity of the paralysis by calculating four ratios. The difference between the affected paralysis side and the unaffected side with various expressions was calculated in [34]. In [35], iris segmentation and LAC-based key point detection were employed to extract the symmetry features in three regions of the face image to quantitatively classify and assess facial paralysis.…”

Classification of facial paralysis based on machine learning techniques

“…The majority of studies in this area are based on datasets of 2D images [6,13] with small number of cases and lack in severity levels variety. As a result, they have limited classification accuracy and hence, are not suitable for large-scale applications.…”

“…Anguraj and Padma [ 13 ] developed a method for classifying the severity level of FP into three categories: mild, moderate, severe beside the normal case. First, Salient Point Selection Algorithm (SPSA) is used to assign a grade for facial movements.…”