Abstract
Facial paralysis (FP), as a highly prevalent neurological dysfunction disease worldwide, has long faced challenges such as strong subjectivity in assessment and difficulty in quantifying therapeutic effects in its clinical diagnosis and treatment. Traditional scales rely on physicians' experience. Neuroelectrophysiological examinations are invasive, while imaging evaluations are costly. The rise of intelligent sensing technology provides a new path to break through these limitations. Intelligent sensing technology has significantly improved the accuracy of FP recognition and assessment through multi-modal data fusion and dynamic monitoring. Its clinical value is not only reflected in the improvement of diagnostic efficiency, but also in promoting a fundamental change in the diagnosis and treatment model of FP. The artificial intelligence-assisted analysis mainly focuses on using machine learning algorithms to conduct in-depth exploration and analysis of the surface electromyogram (sEMG) signals of patients with facial paralysis, the motion trajectory data obtained through three-dimensional (3D) motion capture, as well as the data from patients' self-assessment scales. This study systematically reviews the innovative applications of intelligent sensing technology in the recognition and evaluation of FP, focusing on three major technical directions: sEMG, 3D motion capture, and artificial intelligence assisted analysis.