a brilliant invention that revolutionized early detection
A face scanner for mobile phones that can help doctors detect a stroke in seconds. Biomedical engineers from RMIT University and the State University of São Paulo have developed this AI-powered software. As described, this method significantly exceeds current technologies in accuracy and speed.
The study was led by researcher Guilherme Camargo de Oliveira from RMIT University and the State University of São Paulo, under the guidance of team leader Professor Dinesh Kumar.
“Early detection of stroke is important because timely treatment can significantly improve recovery outcomes, reduce the risk of long-term disability and save lives.”
The main goal of the tool is to detect as early as possible whether a patient is in the post-stroke phase in order to provide information as quickly as possible.
They say the tool has an 82% accuracy rate for detecting stroke, although it was in no way designed to replace full diagnostic tests.
How do you achieve your goal? With facial recognition algorithms and artificial intelligence. Researchers highlight the difficulty of detecting strokes, as sometimes the signs are very subtle. The patterns analyzed by this tool are related to facial expression recognition, facial symmetry analysis, and specific muscle movements known as action units.
The Facial Action Coding System (FACS), first developed in the 1970s, classifies facial movements based on the contraction or relaxation of facial muscles, providing a detailed basis for analyzing facial expressions.
“One of the key things that affects people who have had a stroke is that their facial muscles often become lopsided, so one side of the face behaves differently to the other.”
The goal is to develop a mobile phone app that can not only detect these patterns, but also other neurological conditions that affect facial expression (this is the key to making the tool even more accurate).
Image | RMIT University
In Hatake | Your Fitbit Will Talk to You. Google Improves Its AI to Focus on Health