The Development of a Facial Asymmetry-Based Model for Elderly Stroke Detection

Advanced age brings a heightened vulnerability to strokes, often linked to factors like diminished vascular resilience, high blood pressure, and age-related health conditions. It's important for elderly and their caregivers to be aware of the signs and symptoms of stroke, which can include sudden numbness or weakness in the face, arm, or leg, especially on one side of the body. Facial paralysis constitutes a prevalent acute stroke symptom characterized by impaired facial mobility discernible through speech and non-speech activities. Presently, research endeavors harness machine learning techniques to create automated facial asymmetry assessments; however, limited attention is directed toward the elderly demographic. Our study aims to fill a crucial gap in stroke detection for elders by developing a model based on facial asymmetry analysis. Our approach employs cosine similarity to quantify facial asymmetry between the left and right hemispheres, subsequently utilized as discerning features for stroke detection.