Visual Facial Paralysis Detection using InceptionResNetV2

Facial paralysis, stemming from nerve issues, results in the inability to control facial muscles, leading to asymmetry and weakness. This condition not only affects appearance but also disrupts daily activities. Diagnosis is time-consuming and requires specialized expertise and equipment. To address these challenges, a deep learning-based is proposed to analyze facial expressions and distinguish between normal and paralyzed states. The Facial Paralysis Detection leveraging InceptionResNetV2 by including preprocessing, augmentation, and classification on facial images. Evaluation on dataset of 2000 images of facial paralysis and normal face, these images are divided into training, validation, and testing sets with a ratio of 5:1:1, achieves 92.7% accuracy in distinguishing normal and paralyzed facial expressions, marking significant progress in healthcare diagnostics. Further experimentation will be carried out on alternative deep-learning structures in order to attain the highest precision.