Facial digital markers For hypomimia detection in Parkinson's disease: A systematic review

Hypomimia, or facial masking, is a clinical symptom characterized by decreased facial movement and emotional expressions, commonly seen in individuals with Parkinson's disease (PD). This review provides a comprehensive analysis of the state of the art on automated hypomimia detection in PD. As studying PD through digital facial features is an emerging field, we conducted a broad review of the literature without imposing specific time limits, in order to capture both historical and current approaches to hypomimia analysis. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed to systematically select and evaluate relevant studies. We examine hypomimia-detection approaches across various modalities, namely static images, video sequences, electromyography signals, and optoelectronic systems. Our review proposes several structured categorizations of research studies, based on the scenario type, whether emotional or non-emotional, considered to assess facial muscle movements, the types of facial features extracted, and the computational methods applied for hypomimia analysis, namely statistical tests, machine learning, or deep learning techniques. Additionally, we explore the interpretability of AI models for hypomimia detection, revealing patterns associated with the symptom. Finally, we investigate the link between hypomimia and other clinical symptoms of PD.