Fully Automatic Stroke Symptom Detection Method Based on Facial Features and Moving Hand Differences

Abstract

In this article, we suggest a new approach that automatically analyzes the degree of left and right symmetry of the cheek wrinkle lines and the movement of both arms in order to detect the symptom of the early stroke. To achieve that was created a special model, using Active appearance model (AAM) algorithm that detected cheek wrinkle line by training face dataset. Hand moving part our method was detected combining motion detection algorithm and skin model of the patient. Experimental results show that our method correctly detects stroke warning signs and symptoms to save patients life without wasting time.

Identifiers

DOI

10.1109/ismac.2019.8836166

Citations and references