Invention: AI brings earlier care to new Parkinson's patients
After watching a YouTube clip of Parkinson's patient Michael J. Fox, US-based Erin Smith developed an AI-powered app that uses video to enable early detection of Parkinson's disease. Her intervention can help to slow development of the condition.
When Erin Smith was a child, her mother encouraged an interest in science, even turning the family kitchen into a makeshift laboratory. Later, shortly after starting high school, inquisitive young Smith watched a YouTube video of Michael J. Fox, a high-profile Parkinson's patient who showed the facial expression trait known as "mask face". She began to question whether facial expressions could be analysed to monitor changes in the brain. Smith learned to code and began working on what would later become FacePrint, an AI-powered application that records facial expressions and uses computer vision to accurately detect minute indicators of early-onset Parkinson's.
The computer vision algorithms are trained to interpret video data of Parkinson's patients and healthy people making facial expressions. As a result, FacePrint can detect Parkinson's with 95% accuracy and objectively capture and digitise changes that occur five to ten years before diagnosis based on traditional motor symptoms. Additionally, Smith and her team have focused on building an inclusive dataset, enabling accurate detection across gender and race.
More than 10 million people worldwide live with the condition, with cases expected to increase significantly as the ageing population grows. Earlier diagnosis enables faster treatment, potentially delaying more serious symptoms of Parkinson's, such as tremors and walking difficulties. FacePrint therefore contributes to UN SDG 3 (Good Health and Well-Being) at every stage of life. As the tool is low-cost and can be used remotely by non‑experts, it also addresses UN SDG 10 (Reduced Inequalities).
AI brings earlier care to new Parkinson's patients