A new smartphone app developed by researchers at RMIT University uses AI to detect strokes by analyzing facial asymmetry, particularly the droop in one side of the mouth. This app, intended for use by paramedics and first responders, can identify strokes with an accuracy of 82%, which is comparable to current methods used by medical professionals. The app works by having the patient attempt to smile while the app records and analyzes their facial movements in real-time, ensuring privacy as no video is stored. Early detection is critical as it enables timely treatment, reducing the risk of long-term disability and improving patient outcomes. The tool has shown promising results in pilot studies and aims to be a cost-effective, easily deployable solution to improve stroke diagnosis in high-stress environments and across diverse populations.