A smartphone app has been developed by biomedical engineers to identify symptoms of stroke using an AI-powered facial express analysis.

The app is designed as a tool to assist caregivers and family members in quickly recognizing the typical signs of stroke in real time, giving them the critical opportunity to call for medical help. Patients have a much better chance of surviving and avoiding permanent disability if they receive treatment immediately. Strokes occur due to disrupted blood flow to the brain, and early detection is vital for improving patient outcomes.

The team behind the app is led by Professor Dinesh Kumar and PhD scholar Guilherme Camargo de Oliveira at RMIT University in Australia. By using AI algorithms to examine facial symmetry and detect muscle movements associated with stroke (based on the Facial Action Coding System), the researchers achieved an 82-percent accuracy in detecting stroke symptoms with the app.

As part of the development process, the team analyzed video recordings of facial expressions of both stroke patients and healthy individuals to enhance the app’s effectiveness. The app, however, is not meant to replace traditional clinical assessments but rather to act as a complementary tool for rapid initial evaluation.

The team believes the app could be particularly valuable in diverse patient populations, where stroke symptoms might vary or be overlooked. Kumar cited studies describing how strokes are often missed in emergency settings and community hospitals, hence the need for a user-friendly and reliable smartphone app as a diagnostic aid.

Leave a comment

Your email address will not be published. Required fields are marked *