Guilherme Camargo de Oliveira, a PhD student from the Royal Melbourne University's (RMIT) Corps of Biomedical Engineers and the State University of São Paulo, has developed an AI-based facial screening tool.Healthcare professionals only need to use theSmartphoneIt is possible to determine in a few seconds whether a patient issuffer a paralyzing stroke.
▲ PhD student Guilherme Camargo de Oliveira demonstrates the facial screening tool with RMIT University Visiting Associate Professor Nemuel Daniel Pah (pictured)
This tool in detecting strokeAccuracy of 82%The newest addition to the RMIT Stroke Lab is a comprehensive clinical diagnostic test for stroke, but it will not replace the comprehensive clinical diagnostic test for stroke, and will only be used to help identify people in need of treatment more quickly. Professor Kumar from RMIT's School of Biomedical Engineering says that early detection of stroke and timely treatment can significantly improve recovery outcomes and reduce the risk of long-term disability.
The tool deeply integrates AI with facial recognition technology, theDetecting stroke by analyzing facial symmetry and specific muscle movements (called action units). The tool uses the Facial Action Coding System (FACS), originally developed in the 1970s, to classify facial movements by contraction or relaxation of facial muscles, providing a detailed framework for analyzing facial expressions.
Team member de Oliveira stated thatOne of the key parameters affecting stroke patients is that their facial muscles usually become unilateral in their movements, so the muscles on the two sides of the face behave differently. The tool centers on detecting whether the asymmetry of the face changes when it smiles. The study used videos of facial expressions examined in 14 post-stroke patients and 11 healthy controls.
The team plans to work with healthcare providers toDevelop the tool into a mobile applicationto be able to detect other neurological disorders that affect facial expressions.